close

Вход

Забыли?

вход по аккаунту

?

THE CONSTRUCTION AND STANDARDIZATION OF THE PURDUE MECHANICAL ASSEMBLY TEST

код для вставкиСкачать
DOCTORAL DISSERTATION SERIES
jj UNIVERSITy MICROFILMS
H ANN A I I O I • MICHIGAN
/
O
t
Z
Q
“The results obtained and the thesis prepared in con­
nection with the regularly assigned thesis subject are
the property of the University and no part of the same
may be reproduced or published without the written
consent of the President of the University nor may it
be used, directly or indirectly, in support of or in con­
demnation of any product or procedure referred to
therein.” (Purdue University, Board of Trustees:
Minute VI, Meeting of April 21,1937).
^
PU R D U E UNIVERSITY
T H I S I S T O C E R T IF Y T H A T T H E T H E S I S P R E P A R E D U N D E R MY S U P E R V I S I O N
BY
Llaur3 ce R. Granny
E N T IT L E D
The Construction and Ctandardi^ation of the
Purdue Mechanical Assemblv Test
C O M P L I E S W IT H T H E U N IV E R S IT Y R E G U L A T I O N S O N G R A D U A T IO N T H E S E S
A N D I S A P P R O V E D B Y M E A S F U L F I L L I N G T H IS P A R T O F T H E R E Q U I R E M E N T S
FOR TH E D E G R E E O F
Doctor of Philosophy
P
r o fesso r
H
ead of
S
ix C
habok
chool oh
D
o f Th
e s is
epartm ent
>*£3=^
Z m dfm+
I
T O T H E L I B R A R I A N : ----T H IS T H E S IS I S N O T T O B E R E G A R D E D A S C O N F ID E N T IA L .
B tto a
IR O IH T 1 A B FORM
■!>—
1M
IM CHAJiOK
Abstract*
Maurice R* Graney*
The Construction and Standardization at the
Purdue Mechanical Assembly Test*
May, 1942.
81 pages
Qjbables
21 figures
23 working drawings
29 titles in bibliography*
The purpose of this investigation was to construct a new test of
mechanical ability which did not possess certain faults characteristic
of earlier tests of this kind, and to standardize its administration
and scoring.
The test consists of a series of nine problem boxes of standard
size*
In each box a mechanism may be assembled in such a way that a
mechanical action takes place*
The first problem box contains sub-test
X which is an introductory unit* The eight sub-test units, A-l, A-2,
A—3> A-4, B-l, B-2, B-3, and B-4, which compose the test situation are
divided into two forms of four units each. The order of administration,
in each form, is from the simple to the complex.
The test embodies
certain characteristics, i*e«, sturdy, precision construction, and non—
stereotyped problem situations, which appear to be improvements upon
those tests which were its prototypes.
Complete working drawings show­
ing the construction of each test detail as well as correct test assembly
for each unit have been made.
The eight sub-tests included in the test situation give a suf­
ficient number of assembly tasks to yield a reliability of 0*77*
This was determined by correlating performance on the first four sub­
tests against performance on the second four sub-tests.
In all 338
total test scores were used for this calculation, the number being
made up of 128 college engineering students, 98 machinists, 48
machinists* apprentices, 43 students in a vocational high school, and
21 unclassified adults.
For the first four of these groups, performance
on the test was correlated with some index of success in the activities
in which the groups were engaged,
T^ese indices were grades in shop and
drawing, merit ratings, instructor ratings, and supervisor ratings.
Cor*,
rected coefficients of correlation ranged from 0,20'to 0i51,
A factor analysis of the sub— test intercorrelations revealed the
presence of two factors.
One of these was identified as mechanical in­
sight, the other as mechanical experience.
In these calculations
scores from 203 persons selected from all the groups tested were used.
The general findings of the study indicate that the test may be
too difficult for students in the secondary schools, but should prove
useful in vocational guidance and employee selection on the adult level.
THE CONSTRUCTION ANN STANDARDIZATICK Cl THE
PURDUE I.UvUAjN ICAL ASSEMBLY TEST
A Thesis
Submitted to the Faculty
of
Purdue University
by
Maurice R. Graney
in partial fulfillment of the
requirements for the Degree
of
Doctor of Philosophy
May, 1942
Maurice Richard Graney
Date of birth:
April 1, 1907
Academic career:
B. b. I. E.
Purdue University 1935
Degree conferred *ith distinction
A. b.
Purdue University 1937
Thesis:
An Experimental Analysis of Certain
Learning Reactions in the Rat.
Honary oociety MembershipL
Kappa Delta Pi
Iota Lambda Si0ma
Sigma Ai
Technical Reports and Publications:
The Learning Performance of dpiit e Rats in a Temporal
Maze Pattern Situation.
Co-authored »»itn 0. C. Triable ana presented
at annual meeting of the Indiana academy of
Science, 1935•
A Preliminary- Study of the Delayed Reaction of the Mhite Rat,
Co-suthored rtith 0. C. Trimble and presented at
annual meeting of the i'da«estern Psychological
Association 1936.
Motion Pictures for Industrial Training.
Paper presented at a;inual Conference on rt-rsonnel
an-u Inuustrial Relations, Purdue University 1937*
4 Study Conuucted in Connection with the Drawing Division
of the S . P . £. E . on V isualization.
Parer presented at the annual meeting of the
Society for the Promotion of nntineering Education
IV 3P •
A Or itical nvfluatlon of Placenent Tests in mr^ineerin^
Drawing.
Journal ol mn^ineerint. _^ra<.ing, October 1V3&.
Report on Development of Tests Measuring Certain aspects
of Engineering, Aptltuae.
Co-authare. ..ith Justus Rising an j presented at
annual meeting of the society for the Promotion
of engineering i-aucr tion, 194^ •
Visual aids in the Drafting Room.
Co—authoreu with D. J. Luza ,aer atm presented
at annual meeting of The Indiana Industrial
Education Association, 1941*
Test in- . Devices for the Selection of machine ohop atucfents.
Paoer presented at the annual meeting of the
Indiana Industrial Education Association, 1942,
FOREWORD
An investigation of the type reported herewith is one which
requires much, not only of time, but also of energy and cooperation
on the part of many persons other than the investigator.
instance such cooperation was cheerfully given.
In this
To all of those who
so participated, I wish to -Lender a sincere thanks.
Whatever small
contribution this investigation may make comes as much from them as
from me.
In particular I wish to thank Professor Joseph Tiffin who directed
the study.
Likewise I wish to express my appreciation to Professors
H # H. Remners, E. L. Kelly, C. L. Morgan and Amon Swope, for consider^^
ate advice and criticism of iqy -owrk as the investigation proceeded.
*
Without the help of these persons continuation of the work would have
been unlikely.
I also wish to thank Mr. Theodore Hoffman, Director of Industrial
Relations, Cincinnati Milling Machine Company, Mr. Eugene Maple,
Director of Training, Carnegie-Illinois Steel Corporation, Sheet and
Tin Mill at Gary, Mr. A. P. Johnson, Assistant Personnel Director,
Purdue University, and Dr. Charles Lawshe, Dean of Mechanic Arts High
School, Evansville, for the cooperation which made it possible for me
to contact the subjects of this investigation, and who made available
to me the various criteria of this study.
Also to Dr. Stanley Seashore, Professor Norman Arnold, Mr. Dennis
Price, Mr. John McClure, Mr. Alan Motoux, and my wife, I express appre­
ciation for their invaluable service and advice.
CONTENTS
Page
Historical Background
2.
Description of the Present Test
10
Administration of the Test
19
Results of tne Investigation
25
Summary and Conclusions
39
Bibliography
43
Appendix. A, Test uata for mirciue University
Engineering students
^
Appendix B, Test Data Carnegie-Illinois Steel Corporation
Sheet and Tin i.ill, Gar;y, Indiana
Appendix C, Test Data Cincinnati ..tillingMachine Company
7
12
Appendix D, Test Data, Evansville, Indiana mechanic
Arts High School Students
15
Appendix E, Test Data for Ur.classifiea adults
18
Appendix F, Distributions of sub-test scores
2D
Appendix G, Sub—test Pact, r
2D
lo sis
A ;j oof.'m II, Deb-test Working Drawings
36
1
I
HISTORIC
BACKGROUND
Vocational guidance and employee selection have given rise to the
problem of predicting success in pursuits requiring varying amounts of
mechanical ability.
Numerous tests have been designed to evaluate
certain supposedly critical aspects of this ability.
While these tests
have had their own specific characteristics, they may be grouped into
four rather well defined categories.
These categories imply the
methods of approach employed by previous investigators.
Group one is composed of paper and pencil tests.
One such test,
the Liac^uarrie Test for Mechanical Ability, devised by T. W. MacC^uarrie (15)> is intended to furnish a rough indication of aptitudes
for acquiring manipulative skills.
The test appears to require on
the part of the testee some ability to recognize space relations,
speed and decision of movement, hand-eye coordination, muscular con­
trol and visual acuity.
Another test in this group is O'Rourke's Mechanical Aptitude
Test (18).
It attempts to sample a person's familiarity with certain
tools and their uses.
It assumes that such a familiarity is trace­
able to an aptitude for learning to manipulate these tools.
Similar to part one of O'Rourke's test are the Stenquist Mechani­
cal Aptitude Tests I and II.
Test I is composed entirely of pictorial
material, the task being to associate pictures of related common tools,
mechanical contrivances or parts.
Test II is, in part, similar to
Test I, ana in part, composed of questions about pictures and dia­
grams of machines and machine parts.
These tests are supposed to
indicate an interest in things mechanical and call upon the testee
not only for a knowledge of mechanical relations but also an ability
to reason about them*
Another test composed of pictorial material is G. K. Bennett's
Test of Mechanical Comprehension Form A (2).
This test attempts to
sample a person's knowledge of the way in which mechanical and physical
principles operate.
Emphasis is not placed upon a simple knowledge
of mechanical things, rather it is placed upon the function of mechani­
cal things in every day living.
Also in this group is the Detroit Mechanical Aptitude Examina­
tion for Boys.
This test was devised by Baker and Crockett (1).
The attempt, here, to sample a person's familiarity with things mech­
anical presupposes a relationship between such familiarity ana an
aptitude for mechanical pursuits.
Included in this group, also, are the various form boards devised
for group administration.
Typical of this type of test is the
Minnesota Paper Form Board, a test devised by Paterson, et al., (19)
and patterned after the Geometrical Construction Test in the Army
Beta examination.
The 193^ revision by Likert and’ Quasha (13) contains
sixty—four multiple (five) choice problems each of which requires the
testee mentally to manipulate simple, two-dimensional geometric shapes.
No manual manipulation is permitted.
The test is supposed to measure
the ability to discriminate geometrical patterns.
The second group of mechanical ability tests is composed of tests
which reproduce a job situation in jainature or by analogy.
One of the
earlier tests of this type is the llotorman Test of Uunsterberg (16).
The apparatus designed by Hunsterberg is intended to measure abilities
important in avoiding accidents traceable to street car operators.
It is composed of a diagraraatic representation of a busy thoroughfare
through which the subject might operate a street car.
the person taking the test to
The task is for
ieutify critical situations apparent
in the diagram.
According to Viteles (28) Munsterberg*s Test was modified and
used by Gerhardt (9) who reports satisfactory results.
On the other
hand r’ontegne (7 ) reports negative Tindin- s, as does Rupp (21) when
using a modified form of Kunsterberg's apparatus.
Viteles (28) constructed a L'otorman Selection Test intended to
measure abilities differentiating the safe from the accident-prone
motorman.
The apparatus included four main parts (1), signal board
for visual stimuli; (2), signal board for auditory stimuli; (3),
reaction stand; and, (A), a distraction signal.
the subject must undergo a period of training.
Before being tested,
Viteles (28),
dhello.i (23), arm Lewnurst (6) report satisfactory results from the
use of this test.
Another type of test in tnis group is tue Two-Hand Test aevisea
by Po^de as reported by Viteles (28, pp. 279) and Bingham (3 pp. 135).
The apparatus is composed of a small table so mounted that it can be
moved in any horizontal direction by turning two screws at right
angles to each other.
A paper mounted on this table top contacts a
pencil hela in a fixed position.
made uoon the paper.
By turning the screws, a tracing is
The testee performs the task of making a tracing,
keeping the pencil line within narrow limits.
Score is in terms of
time required and errors co;.uni11ed.
The ability sampled by this test
is supposed to be symptomatic of aptitude for certain trades, such
as that of lathe operator.
Patten. (20) developed tne 'Wisconsin ilinature Test for ikigine
Latne Aptitude, a test similar in principle to the Two-Hand Test.
This apparatus is wired to yield, in part, an electrically obtained
objective evaluation of performance.
The third group of mechanical ability tests is comj>osed of those
tests known as ..ork Samples.
O ’Connor (17)•
Among these tests are those devised by
Apparently the aim of O'Connor's investigations was the
development of a series of tests, the successful performance of -which
would be a measure of aptitudes.
Used singly or in combination, test
scores may be used to predict chances of success in such areas as
box-naking, millwrighting, medical -work, clerical work, etc.
Those
tests intended to be a measure of some phase of mechanical ability
include ,/orksamples Mo. 5, No. 16, No. 17, and No. 75.
worksample
No. 5, known as the "wiggly blocks”, is a set of nine blocks jig-sawed
from a rectangular solia of wood.
The score is given in terms of-
time required by a testee to assemble the nine scattered blocks into
the original solid.
worksample No. 16, interned to measure finger
dexterity, presents the testee ..ith a peg boaru containing luQ holes
ana a tray containing
pegs.
three at a tine, in the holes.
The person tested must p-lace tne pegs,
The score is the tl..e required.
Some­
what akin to this is dorksample No. 17, & test of tweezer dexterity.
In this test tweezers instead of fingers are used to pick up, the pins.
Works ample 75 is composed of a form board from which have been
jig-sawed pieces in a series of abstract designs.
is to reassei.ible the design,
The subject’s task
score is in terns of the tine required.
In this third group are the worksamples developed by Coover (4).
The first of these is a device for measuring deviation in filing.
The testee strokes a file across a metal block.
Deviations from a
truly horizontal stroke, etc., are gauged objectively on a clock—hand
registering attachment.
Other worksamples developed by Coover are a
test of hammering, a test of scaling, and a registering try square and
angle meter.
The fourth group of mechanical ability tests includes those tests
which attempt to sample a person’s oer for nuance in mechanical problem
situations.
..hile these tests do not have a clear homogeneity, they
each uemand on the part of the testee a degree of some abstract ability—
ability to visualize, an ability to note spatial relationships, etc.—
as well as a modicum of manual skill.
One of the earliest of such tests was submitted by ^tenquist (24).
This test, kno*n as the dten uist Lechanical .issembiing lest, measures
a person's ability to put together the tarts of mechanical uevices,
such as a mouse trap, a bicycle bell, etc.
Patterned after, and
supposedly an improvement upon this test is the minnesota Mechanical
Assembly Test of Paterson, et al., (19)•
it is composed of thirty-
three disassembled mechanical contrivances, for each of «hich a
maximum allowable assembly time is specified,
an individual's per­
formance on this test is not to be interpreted as a pure measure of
mechanical intelligence, but as determined, in part, by the individual's
familiarity with mechanical devices.
This fact operates to make the
test a less reliable measuring instrument for adults than for boys.
Paterson, et al. (19) have also developed the Minnesota Spatial
Relations Test, a modification of Link’s (14) Form Board.
is composed of four standard form-boards from each
cut 58 pieces of different forms and sizes.
This test
>.hich have been
The problem is to put each
of the pieces in its proper place.
Included in this group also is the Institute of educational
Research Assembly Test for Girls devised by Toops (25) and revised
by Bun and Metcalf (3 pp. 290).
It is composed of four sub-tests
involving such tasks as insertion of tape, assembling a chain of clips,
paper cutting, and the like.
It «as constructed "with the aim of
duplicating with different materials the test situations presented
to boys in the btenquist Assembling Tests." (8 pp. 58).
At least two investigators have devised tests not readily classifiable in the four above categories.
Rupp (2fc) in selecting
apprentices in the Simens-Schuckert Yorks, Berlin, used a series
of 18 tests,
,/ith these he attempted to measure visual discrimination,
spatial perception, technical comprehension, ana manual ability.
Cox (5) developed a series of tests to aid in securing scientific
evidence based on valid objective criteria concerning the relation­
ship between mechanical ability and other abilities.
four kinds^
The tests are of
mechanical models, mechanical completion, mechanical
explanation, and mechanical diagrams.
The models "were so devised
that the subject could only see the first and last link in the series
of mechanical
hand).
events which occurred when the model was worked (by
He was required to indicate by a simple sketch (with the
addition of such words as he thought necessary) how the observed
movements were brought about" (5 pp. 52).
Paterson, et al. (19) have also developed the Minnesota Spatial
Relations Test, a modification of Link's (14) Form Board.
This test
is composed of four standard form—boards from each o^ ».hich have been
cut 58 pieces of different forms and sizes.
The problem is to put each
of the pieces in its proper place.
Included in this group also is the Institute of Educational
Research Assembly Test for Girls devised by Toops (25) and revised
by Bun and Metcalf (3 pp. 290).
It is composed of._f.our sub—tests
involving such tasks as insertion of tape, assembling a chain of clips,
paper cutting, and the like.
It <vas constructed "with the aim of
duplicating with different materials the test situations presented
to boys in the Stenquist Assembling Tests." (8 pp. 58).
At least two investigators have devised tests not readily
classifiable in the four above categories.
Rupp (2£) in selecting
apprentices in the Simens-Schuckert Arorks, Berlin, used a series
of 18 tests.
V/ith these he attempted to measure visual discrimination,
spatial perception, technical comprehension, ana manual ability.
Cox (5) developed a series of tests to aid in securing scientific
evidence based on valid objective criteria concerning the relation­
ship between mechanical ability and other abilities.
four kinds^
The tests are of
mechanical models, mechanical completion, mechanical
explanation, and mechanical diagrams.
The models "were so devised
that the subject could only see the first and last link in the series
of mechanical
hand).
events which occurred when the model was worked (by
He was required to indicate by a simple sketch (with the
addition of such words as he thought necessary) how the observed
movements were brought about" (5 pp. 52).
There are numerous other tests, either performance such as the
Stanford Motor Skills Unit (22), or paper and pencil, such as the
Purdue Vocational Achievement Tests (29), which sample abilities related
to mechanical ability.
Tests of this type, as a rule, measure a more
limited activity than is generally accredited mechanical.
For this
reason, plus the fact that their originators do not claim that the
tests measure mechanical ability, discussion of them is omitted here.
Investigators have not agreed in ansv*ering the question about
the nature of mechanical ability.
One point of view, that mechanical ability is a single factor
common to all kinds of manipulative operations, whether manual or
abstract, has considerable support.
of view.
Stenquist (21) supports this point
He cites the fact that mechanical tests intercorrelate to
the extent of about 0.6 and 0.7 on the average.
On the whole he
concludes that there is a general mechanical aptitude, “general in
the sense that it does not pertain to any special trade, and mechani­
cal as is more or less obvious from its nature.11 (21 pp. 251) •
gives further support to this concept.
Cox (5)
The results of his investiga­
tions, he holds, point to the existence of "(l) the general factor, g;
(2) the group factor, mj and (3) one or more factors 'specific* to
the particular 'mechanical* task in question". (5 pp» 185 )•
A different point of view is presented by Paterson, et al. (19)*
Their investigations failed to reveal a mechanical ability group factor.
They conclude that "Analysis of the organization of mechanical ability
indicates that it probably does not involve any single general factor.
Low intercorrelations between different measures of mechanical ability
suggest, that factors of high specificity play a major role.” (19 pp.
300.)
vThether or not this point of view is distinctly different from
that of Cox is discussed by Harvey (10).
Harvey points to a difference
in the connotation of terms used, as well as a difference in general
orientation of the investigations, as a possible explanation of differ­
ent results.
Coover (4) supports the position of specific as opposed
to group factors in mechanical ability.
He stresses the importance
of an operational definition of the term and postulates a hierarchy
of specific factors.
He points out that some specific factors may be
combined to form a quasi-group factor, but claims that this is not
pertinent to the problem.
Viteles (28) mentions the work of Penin, Muscio and Seashore in
support of the idea that most mechanical tasks are essentially differ­
ent from each other and consequently call for a specific type of
mechanical ability.
It is difficult to extract a general conclusion concerning
the
nature of mechanical ability from conclusions so markedly different.
It is not surprising, however, that such different conclusions vo^re
reached when a comparison is made of the kinds of tests used,
dten-
quist ana Cox, on one hand, used tests which ueal primarily with mental
comprehension or understanding of mechanical principles, whereas
Seashore and Coover, for example, used tests of muscular coordination
or dexterity.
Perhaps it is in accordance with the various findings
to say that in the comprehension of mechanical principles there seems
to be considerable evidence for the existence of a general factor, but
in the manipulative aspects of muscular performance, the evidence is
against any theory of general motor ability.
The problem of general or specific factors in mechanical ability
may be obviated by viewing it in a different way, namely from the
Gestalt position*
Here the attention is centered upon the pattern
of an activity rather than upon its parts.
Any mechanical activity
may be classed as a mosaic of simpler elements.
Such elements, as
Juhasg (12), accoraing to Viteles (28 pp.237) points out, can be
isolated and examined.
"The difference between two acts of skill,
such as drill press and automatic screw machine operation, lies only
in the kind or quantity of traits involved, and the mechanical ability
(or abilities) underlying them can be distinquished as stable and
unchangeable factors— •"
10
II
DESCRIPTION OF PRESENT TEST
The Purdue Mechanical Assembly Test consists of a series of nine
problem boxes of standard sizes*
In each box a mechanism may be
assembled in such a way that a mechanical action takes place.
The
first problem, box contains sub—test X. which is an introductory unit*
The person tested is shown this illustrative problem first, and is
instructed regarding the nature of the tasks which are to follow.
Problem boxes A-l, A-2, A-3, A-4, B-l, B-2, B-3, and B-4, contain the
eight sub-tests which constitute the complete test.
The person tested
is allowed a certain maximum tine to assemble the mechanism in each of
these eight boxes.
The test involves certain changes over previous mechanical
assembly tests.
The Stenquist Mechanical Assembling Test and the
Minnesota Mechanical Assembly Test represent the good efforts to
measure mechanical ability with a single test.
some weaknesses.
They possess, however,
First, the fact that they employ as test problems
contrivances selected from the workaday world serves to reduce their
effectiveness.
As is pointed out by Bingham, an individual's per­
formance cannot be interpreted as a "pure measure of his mechanical
intelligence, but that it must be appraised in the light of what is
known about his previous background...." (3 PP» 308).
This fact
operates to make these tests less valid measuring instruments for
adults than for boys.
Second, the workmanship employed in the construc­
tion of the contrivances used is not satisfactory.
In general, five
cent and ten cent mechanical devices cannot stand much abuse.
This
fact operates to impair the standardization of conditions under which
11
the test must be given.
As an example, consider the light guage wire
used on a five cent mouse trap.
assembling the parts.
This easily may be bent by a testee in
The subject next to follow then deals with a
test object which is no longer standard.
The time consumed by this
subject will not be a true measure of the time he would require to
assemble the standard test object.
The Purdue Mechanical Assembly Test was constructed along the
broad outlines 01 the two tests criticized above, but with the hope of
eliminating the defects cited.
It is composed of nine sub—tests, each
of which, -.hen assembled, embodies a certain mechanical principle or
combination of principles,
Each sub—test is dynamic.
Vhen it is
properly assembled, it can be operated in its own unique way.
However,
each sub—test is a mechanism not employed in the workaday world, and is
one constructed of quality material, sturdy and strong, by skilled
craftsmen.
The elimination of stereotyped mechanical contrivances, in favor
of new mechanical problem situations, rules out the effect of chance
familiarity with the test at hand for all persons, regardless of age
or experience.
Nevertheless, the fact that each sub-test employs in
its assembly standard mechanical items such as levers, links, cams,
gears, etc., tends to make the solution simpler for experienced as
opposed to mechanically naive laymen.
The principles of assembly
in each test also are drawn from standard design practices, in most
instances.
Linkages, gear trains, the rack and pinion, are all
employed.
Each sub—test is built of parts which are relatively large and
strong.
While this tends to increase the test in bulk and weight, and
12
perhaps cost, it likewise tends to keep the test in practically the
same condition for all testees.
One set of sub—tests, for example,
was carefully scrutinized after administration 100 times.
No observ­
able change from its original condition was evident.
In general, the administration of the test is the same as that
employed in the Stenquist Mechanical Assembling Tests and in the
Minnesota Mechanical Assembly Test.
After a standardized instruction
is given to him, the subject is presented with the unassembled subtests.
Score is in terras of time required to make all of the assemblies.
This test, however, introduces a unique factor:
each sub-test is
assembled in a box of standard size.
The eight sub—tests included in the test situation give a suf­
ficient number of assembly tasks to yield a reliability of 0.77.
This was determined by correlating performance on the first four sub­
tests against performance on the second four sub—tests.
In all 338
total test scores were used for this calculation, the number being
made up of 128 college engineering students, 98 machinists, 48
machinists* apprentices, 43 students in a vocational high school, and
U
21 undHssified adults.
For the first four of these groups, performance
on the test was correlated with some index of success in the activities
in which the groups were engaged.
Pearson correlation coefficients
showing the various relationships are given in Table I.
**• more de­
tailed report of these investigations is presented in Chapter IV of
this study.
Table I.
Correlations with Indices of Success.
Group
Criterion
Engineering Students
Grades in drawing, shop
Correlation
Coefficient
(Corrected)
0.43
descriptive geometry
Machinists
Merit rating
0.51
Machinists’ Apprentices Xnstructor's ratings
0.34
Vocational High
0.20
Instructor's ratings
School Students
The nine sub-tests include an introductor unit, test M; four
units in Form A, A—1, A-2, A-3, A-4j and four units in Form B, B-l,
B—2, B-3, B—4.
Fig. 1.
The parts of each sub-test are presented on a mountin
Parts of sub-test A-l placed on Mounting Board.
Fig. 2.
board.
Parts of' sub—test A-2 placed on Mounting Board,
This is a piece of bakelite twelve inches square and three—
h
eights of an inch thick on which are steel pins or in which are cut
slots for the proper placing of sub—test parts.
By means of such
mounting board, each subject is introduced to the unassembled test
in the same fashion.
All separate parts are chrome plated.
The parts
are assembled in bakelite boxes each nine inches by six inches in plan
ana of a height great enough to accomodate the completely assembled
mechanism.
The bottom surface of each box is a one—quarter inch
aluminum plate.
Fastened to this plate ere bearings, pivot pins,
etc., upon which the assembled mechanism rests.
not the same for al1 sub-tests.
Obviously these are
Also the slots or holes necessary to
accomodate projecting parts are cut in the sides of the bakelite box.
Each box has a bakelite cover.
The parts of each assembly, placed on the mounting boards in the
positions they assume in the test situation are shown in Figures 1 to 8.
Fig. 3.
Parts of sub-test. A-3 placed on Mounting Board.
Fig. 4.
Parts of sub-test A-4 placed on Mounting Board
Fig. 5*
Parts of Sub-test B-l placed on Mounting Board.
Fig. 6.
Parts of sub-test B-2 placed on Mounting Board.
Fig. 7*
Parts of sub-test B-3 placed on Mounting Board
Fig. 8.
Parts of sub-test B-4 placed on Mounting Board.
Fig. 9.
Fora A and Form B sub-tests assembled.
Y.rhen any sub—test is assembled, it may be operated manually from,
the outside of the box; the operating crank or push-bar always pro­
jecting through.
Origination of the mechanism's movement is accomplish­
ed in one of t»o ways.
For some units this is done by turning a crank
in a clockwise direction, for others by pushing a push-bar in and out.
A testee's attention is drawn to this fact when he is.given the test
instructions•
Identification of a correct assembly is made in two ways.
the mechanism operates freely and smoothly.
First,
Second, a sliding bar which
protrudes through a slot in the side of the box moves with a recipro­
cating action.
These conditions prevail for all sub-tests.
The design
in all cases is such that the parts may be fitted together in only one
way and still permit these two conditions.
Operation of each sub-test
may take place when the box cover is in place.
19
III
ADMINISTRATION OF THE TEST
The first step in the administration of the test is to acquaint the
subject with the test situation.
This is done by using the introductory
sub-test X, and the printed ‘instructions for administering the test*
The subject is seated at a table beside the examiner and shown sub—test
X completely assembled.
After filling in the headings on the Individual
• PURDUE MECHANICAL ASSEMBLY TEST
Individual Record Sheet
Name
Subject No.
Age (nearest year)___________
Total Time ________________
2-Score
Sub-test
Date
Hour
Order
Elapsed Time
Introductory
Form A
1
2
3
4
Total
Form B
1
2
3
U
Total
Remarks:
Fig. 10.
Individual Record Sheet
2-Score
20
Record Sheet, shown in Figure 10, the examiner proceeds with the
instructions, virhich are as follows:
»»Pur due Mechanical Assembly Test.
(Read these instructions aloud to all persons taking test.)
This is a test of mechanical insight.
It is composed of nine
units, each of which, when assembled, embodies a certain mechanical
principle or combination of principles.
Each unit, when properly
assembled, can be operated in its own unique way.
The first unit is introductory.
Here it is. (Present unit X).
It is already assembled and operates this way. (Demonstrate).
principle illustrated here is the lever.
The
Observe that when I push
this bar, (Push), the opposite bar moves.
I am going to ^.ke this assembly apart.
Watch carefully, for you
are to put it back together as it is now, and I will time you as you
do it.
(Dissemble the unit).
questions?
Now, there you are.
(Answer questions, if any.)
Do you have any
All right, you assemble the
parts as they were in the first place.
(Subject assembles unit X.
Record time).
Now you have an idea of the v*ay in which this test is worked.
All of the other units are like this one in one respect:
they
operate from the outside of the box by either cranking or pushing,
as in this problem, never by pulling, lifting, or moving sidewise, etc.
Always either push or crank.
The resulting action will always be a
bar sliding in and out such as this (operate unit X).
All of the other units differ from this one in one respect: the
parts are unassembled.
You are given the parts and the box.
The
problem is to assemble the parts correctly as quickly as you can.
21
Remember, (1), no tools are required to make any assembly; (2)
all parts fit together easily and the mechanism operates easily — if
you have to force anything, your assembly is incorrect.
The remaining eight units are divided into two groups, Group A
and Group B, each having four units.
each group.
You are allowed 50 minutes for
The first unit is relatively simple.
five minutes to assemble it.
For the second you are allowed 10
minutes; the third 15; and the fourth 20.
of minutes you take to assemble all units.
if you care to.
You are allowed
Your score is the number
You may rest between units
The time required to assemble a unit starts when you
open the box for that unit.
Do you have any questions? (Answer, if any).
Remember, it will never be necessary to force any of the parts
into place, nor to force the assembled unit to operate.
You -uay now start on unit A-l (or b-1) •
(Order of units: A-l, A-2, A-3, A-4; B-1, B-2, B-3» B-4.)
(Alternate order: B-1, B-2, B-3, B-4; A-l, A-2, A-3, A-4.)"
A subject may study the parts mounted on the mounting board and
the box into which the parts are assembled for as long as he wishes.
However, he is not permitted to touch any of the parts, nor to remove
the lid from the box until he is reacay to start.
As the subject lifts
the lid from the box, the examiner starts a stop-watch.
Time is called
by the subject when he has completed the assembly and found that it
operates smoothly.
The examiner records the time to the hundredth of a minute.
In
event the subject fails to complete the assembly within the allotted
time, (see Table II) the examiner should complete it for him.
He thus
obtains whatever benefit nay be derived from seeing the problem com­
pleted as do those who complete the problems themselves*
Table II.
Test Time Allotment.
Sub—Test
Maximum Allowable Time
X
No limit
A-l
5 minutes
A-2
10 minutes
A-3
15 minutes
A-4
20 minutes
Form A
50 minutes
B-1
5 minutes
B-2
10 minutes
B-3
15 minutes
B—4
20 minutes
Form B
50 minutes
Total
100 minutes
A subject may start with sub-test A—1 first and proceed with
Form A followed by Form B, or he may start with sub-test B-1, assemble
the units in Form B before the units in Form A.
Such a procedure makes
it possible for an examiner to test two subjects simultaneously.
It is
recommended that he use two stop-watches if this procedure is followed.
Summation of sub-test times on units A-l, A—2, A-3 and A—4
rounded to the nearest half minute yields the score for Form A.
Like­
wise the rounded sura of sub—test times on units B-1, B—2, B—3 and B—4
gives the score for Form B.
Adding the scores for Form A and Form B
gives a total for the entire test.
As may be observed, the maximum
score for the whole test is 100 minutes.
The best performance to
23
date yielded a score of 23• 5 minutes.
However, the minimum assembly
time for the eight sub—tests in Form A and Form B, not allowing time
to s tudy the solution, is approximately 12 minutes.
Scores made by those subjects assembling Form A before Form B
are not strictly comparable with scores made by subjects assembling
Form B first.
The means, standard deviations, and standard deviations
of the means on each form for 203 subjects taking Form A first and 135
subjects taking Form B first are presented in Table III.
Table III.
Form A and Form B scores under the two different
conditions.
Form A scores (11=203)
Form A first
Mean
Sigma
Sigma of mean
Fori- B first
Form B Scores (N=135)
Form A first
Form B first
35.5
min.
32.1 min.
29.3
min.
33.6 min.
9.1
min.
12.6 min.
6.1
min.
8.1 min.
0.78 min.
1.1 min.
0.42 min.
0.70 min.
It may be observed from Table III that there is a difference
in mean score between Form A scores when Form A is first and when
second; and also a difference in mean score between Form B scores when
Form A is first and when second.
To make it possible to compare the
performance of individuals in the two categories, all scores are con­
verted into Z-scores.
These conversion tables are presented on
pages 23 a, 23 b, 23 c, 23 d.
Figure 11 shows a typical test situation with the. subject assem­
bling the parts of sub—test A-2.
Figures 12 through 20 show projection drawings of the nine sub­
tests after assembly.
23 a
Part I. Form A totals when Form A is first.
r Scores
Z—Scores
Raw Scores
Haw Scores
Z-Scores
Z-Scores
50
-1.59
36.5
-0.11
23
1.37
49.5
-1.54
36
-0.05
22.5
1.43
49
-1.48
35.5
0.0
22
1.48
46.5
-1.43
35
0.05
21.5
1.54
48
-1.37
34.5
0.11
21
1.59
47.5
-1.32
34
0.16
20.5
1.64
47
- 1.26
33.5
0.22
20
1.70
46.5
- 1.21
33
0.27
19.5
1.75
46
-1.15
32.5
0.33
19
1.81
45.5
- 1.10
32
0.36
16.5
1.86
45
-1.04
31.5
0.44
16
1.92
44.5
-0.99
31
0.49
17.5
1.97
44
-0.93
30.5
0.55
17
2.03
43-5
- 0.88
30
0.60
16.5
2.08
43
0.82
29.5
0.66
16
2.14
42.5
-0.77
29
0.71
15.5
2.19
42
-0.71
28.5
0.77
15
2.25
41.5
-0.66
28
0.82
14.5
2.30
a
-0.60
27.5
0.86
14
2.36
40.5
-0.55
27
0.93
13.5
2.41
40
-0.49
26.5
0.99
13
2.47
39.5
—0.44
26
1.04
12.5
2.52
39
-0.38
25.5
1.10
12
2.58
38.5
-0.33
25
1.15
11.5
2.63
38
-0.27
24.5
1.21
11
2.69
37.5
-0.22
24
1.26
10.5
2.74
37
-0.16
23.5
1.32
10
2.80
23b %
Part II.
■ Scores
Form A scores when Form B is first.
Z—Scores
Raw Scores
Z-Scores
50
-1.42
36.5
-0.35
23
0.72
49.5
-1.36
36
-0.31
22.5
0.76
49
-1.34
35.5
-0.27
22
0.60
48.5
-1.30
35
-0.23
21.5
0.84
48
-1.26
34.5
-0.19
21
0.88
47.5
-1.22
34
-0.15
20.5
0.92
47
-1.18
33.5
-0.11
20
0.96
46.5
-1.14
33
-0.07
19.5
1.00
46
-1.10
32.5
-0.03
19
1.04
45.5
-1.06
32
0.01
18.5
1.08
45
-1.02
31.5
0.05
18
1.12
44.5
-0.99
31
0.09
17.5
1.16
44
-0.95
30.5
0.13
17
1.20
43.5
-0.91
30
0.17
16.5
1.24
43
-0.87
29.5
0.21
16
1.28
42.5
-0.83
29
0.25
15.5
1.32
42
-0.79
28.5
0.29
15
1.35
41.5
-0.75
28
0.33
14.5
1.39
41
-0.71
27.5
0.37
14
1.43
40.5
-0.67
27
0.41
13.5
1.47
40
—0.63
26.5
0.45
13
1.51
39.5
-0.59
26
0.49
12.5
1.55
39
-0.55
25.5
0.53
12
1.59
38.5
-0.51
25
0.56
11.5
1.63
38.
-0.47
24.5
0.60
11
1.67
37.5
-0.43
24
0.64
10.5
1.71
37
-0.39
23.5
0.68
10
1.75
Raw Scores
Z-Scores
23 c
Part III.
Raw Scores
Form B scores when Form A is first.
Z-Scores
Raw Scores
Z—Scores
Raw Scores
Z-Scores
50
-3.42
36.5
-1.19
23
1.04
49.5
-3.34
36
-1.11
22.5
1.12
49
-3.2o
35.5
-1.03
22
1.20
48.5
-3.17
35
-0.94
21.5
1.28
48
-3.09
34.5
-0.86
21
1.37
47.5
-3.01
34
-0.78
20.5
1.45
47
-2.92
33.5
—0.70
20
1.53
46.5
-2.84
33
-0.61
19.5
1.61
46
-2.76
32.5
-0.53
19
1.70
45.5
-2.68
32
-0.45
18.5
1.78
45
-2.59
31.5
-0.37
18
1.86
44.5
-2.51
31
-0.28
17.5
1.94
44
-2.42
3^.5
-0.20
17
2.03
43.5
-2.35
30
-0.12
16.5
2.11
43
-2.26
29.5
— 0.04
16
2.19
42.5
-2.18
29
0.05
15.5
2.27
42
-2.10
28.5
0.13
15
2.36
41.5
-2.02
28
0.21
14.5
2.44
41
-1.93
27.5
0.29
U
2.52
40.5
-1.85
27
0.38
13.5
2.6o
40
-1.77
26.5
0.46
13
2.69
39.5
-1.69
26
0.54
12.5
2.77
39
-1.60
25.5
0.62
12
2.85
38.5
- 1.52
25
0.71
11.5
2.94
38
-1.44
24.5
0.79
11
3.04
37.5
-1.36
24
0.87
10.5
3.12
37
-1.27
23.5
0.95
10
3.20
n,
23d
Part IV.
Raw Scores
Form B Scores when Form B is first.
Z-Scores
Raw Scores
Z-Scores
Raw Scores
Z-Scores
50
-2.01
36.5
-0.36
23
1.31
49.5
-1.95
36
-0.29
22.5
1.37
49
-1.89
35.5
-0.23
22
1.43
48.5
-1.83
35
-0.17
21.5
1.49
48
-1.77
34.5
-0.11
21
1.55
47.5
-1.71
34
-0.05
20.5
1.61
47
-1.65
33.5
0.01
20
1.68
46.5
-1.58
33
0.07
19.5
1.74
46
-1.52
32.5
0.14
19
1.80
45.5
-1.46
32
0.20
18.5
1.66
45
-1.40
31.5
0.26
18
1.92
44.5
-1.34
31
0.32
17.5
1.98
44
-1.28
30.5
0.38
17
2.05
43.5
- 1.22
30
0.44
16.5
2.11
43
- 1.22
29.5
0.50
16
2.17
42.5
-1.09
29
0.57
15.5
2.23
42
- 1.03
28.5
u .63
15
2.29
41.5
-0.97
26
0.69
14.5
2.35
41
-0.91
27.5
0.75
14
2.41
40.5
-0.85
27
0.81
13.5
2.48
40
-0.79
26.5
0.87
13
2.54
39.5
-0.72
26
0.94
12.5
2.60
39
-0.66
25.5
1.00
12
2.66
38.5
- 0.60
25
1.06
11.5
2.72
38
-0.54
24.5
1.12
11
2.78
37.5
—0 •48
24
1.18
10.5
2.84
37
-0.42
23.5
1.24
10
2.90
24.
Fig. 11.
Testee assembling sub—test A—2.
...
'/////,
FIG. 12. SUB-TEST
X ASSEMB1X
C
,vf,
'-/ 1
o
l
s
}
c
\N
\\
v
s
i
@ AM
/ \ 0
.Y
I
>0/
<■'//////^ .-://A v /s:////M //A Z w ///////.///./,, //A v/ zao//,
rrr
> , v . v * >;&>■>
FIG.I3. S U B -T E S T
A -l
ASSEMBLY
E 3
FIG. 14. SUB-TEST A-2 ASSEMBLY
'—?vS
Ti ii^zr777'
/ v /
'
7 2
\\■
\\
$>
\
k\
\A
ti\\
.
\
k
v\
\
t—
f o i
• '
/
f t> >
/ f )? 7
///'-/'////////A
■
«-. . ■ I<i / { <■/ / ( I
FIG. 15.
TTpyv >} ,rTT?
SUB-TEST A-3 ASSEMBLY
>"rr
• / ./ y
H
8P
P
*
*
i
i
c m —
)
Os
M
n
I
a
\x
I :
'7/ / * ? f*?
/ ■''VV ■■'
.
F tG .1 V
:^
E3
JESSL
\
\w/////AW m//// m
SU B -T E ST
7
B-1 ASSEMBLE
777
N
\N
□
^
rV
"i
W /
777
I
FIG. 17.
SU B-TEST B -2 ASSEMBLY
u
Cl
□
(
OT7
FIG. 19.
SUB-TEST
B-3 ASSEMBLY
///T/7777
<3
, ,,,
T,r ,^.r
, ry,y ,,
-r y j
'
-
,.Z .■
*'■■*..' r ' .
M *
<■■.
x
7 7777 "r/v
< < ,1. . ’ < ,«., * - t . « - « - 4
*
K^
S ^ L L
-rea.
JJJJJ
m
T73
V 7/4/r /. <■/^/</i/, ///,
/ / A 'k.
%-i.j
FIG. 20.
SUB-TEST B-4 ASSEMBLY
•|
f c
■MaMaaaMlMM
m
RESULTS OF THE INVESTIGATION
In the present investigation five groups totaling 338 persons
were studied.
These groups were made up of the following personnel:
Group A
128 Purdue University Engineering
Students.
Group E
98 Machinists from a large midwestern
steel mill.
Group C
48 Apprentice machinists in a large
midwestern machine tool industry.
G
Group D
43 Students in a vocational high school.
Group E
21 Unclassified adults.
A complete record of all scores made by the subjects in these groups
is presented in appendices a, B, C, D, and E.
A statistical analysis of the findings of the investigation
yielded the following:
1.
Reliability.
The reliability of the Purdue Mechanical
Assembly Test was found by correlating Z-scores on Form A with Z-scores
on Form B.
For the total of 338 persons included in the five groups
identified above, the raw correlation coefficient ..as c.62.
When this
was corrected for the total test length by the Spearman-Brown Prophecy
Formula, the coefficient was p.77.
The standard error was-^t 0.02.
A reliability as high as this indicates that the test is an
adequate measuring instrument when used to classify groups of individ­
uals.
Whether or not it is adequate for individual measurement is a
point open to some question.
As Bingham points out, '’Most tests which
26
have value in estimating an individual's aptitudes have reliabilities
within the range from .85 to .97." (3, pp. 261)
Judged by this stand­
ard the present test would be inadequate for individual prediction.
However, Bingham states elsewhere (3» PP. 214) that the reliability of
a measuring device is dependent upon the homogeneity of the group
tested.
A test is likely to have a higher reliability on a heterogen­
ous population than on one which is relatively homogeneous.
The group
of 338 persons included in the present study is heavily weighted with
individuals who, upon a priori judgment, possess a higher than average
mechanical ability,
machinists, machinist's apprentices, engineering
students, and vocational high school students do not represent a random
sanu le of the general population.
It is logical to suppose that a
reliability coefficient determined from a study of a more heterogeneous
population than the present one would be within the range of from 0.85
to 0.97 which Bingham mentions.
A study of this kind is recoiumended
for additional research on the Purdue Mechanical Assembly Test.
2.
Measures of validity.
(a).
Purdue University Engineering students.
It is logical to assume that success in an engineering
curriculum presupposes some measure of mechanical ability.
For this
reason the 128 students listed in Group A, above, were tested on the
present test during the first semester of their college career.
Two
years later cumulative scholarship indices for 119 of these students,
some of whom were not in school, the same length of time as others,
were correlated with the test scores.
-0.02 * 0.09.
The resulting coefficient was
Such an obvious lack of relationship indicates that
success in the first two years of engineering training is determined
27
■to a large extent, by factors inaependent of what is measured by the
Purdue Mechanical Assembly Test.
This is not surprising when the
course content for the first two years is analyzed.
The subject matter
pursued by most students is as follows:
9 hours
(3 courses)
Math
18 hours
(4 courses)
Chemistry
14 hours
(4 courses)
English
(2
Descriptive
Geometry
2 hours
(1 course)
Shop
4 hours
(2 courses)
Surveying
2 hours
(1 course)
Physics
8 hours
(2 courses)
Applied
Mechanics
4 hours
(1 course)
Military
Training
(A
<D
0
o
(0
4 hours
%
Engineering
Drawing
6 2/3
hours
(4 courses)
Technical Courses Variable
Variable
Non—Technical
Courses
Variable
Variable
It may be readily observed that the curriculum for these first
two years is weighted with courses not calling for mechanical ability.
However, five of the courses listed, in the fields of engineering
drawing, descriptive geometry, and shop, are definitely mechanical in
nature.
Using a limited scholarship. index determined by grades in ■
these courses, and correlating with scores on the present test, the raw
correlation coefficient v*as found to be 0.32.
Ahen this was corrected
for attenuation, the coefficient became 0.43 d 0.08.
(The reliability
28
for semester grades was not known.
However, the writer has computed
the correlation between semester grades in drawing and shop courses
on several previous occasions.
0.65 to 0.75.
Such coefficients usually range from
a reliability coefficient of 0.70 for the scholarship
indices was used in the present correction for attenuation.)
This value yields a co«»efficient of determination of 0.18.
Thus, it may be said that, insofar as the group measured is concerned,
the variability in performance in shop, drawing, and descriptive
geometry was about 18 per cent due to whatever ability is measured by
the cresent test.
(b) Machinists' apprentices.
Mix instructors of the 4S apprentices, listed in Group
C above, combined their estimates of the apprentices and gave each a
letter grade.
gories.
These grades ranged from a to D— anu fell in 10 cate­
Correlating such ratings with scores on the present test
yielded a raw correlation of
0
.2
1
tion the resulting coefficient is
.
.hen this is corrected for attenua­
0.34
+ o.ly.
(ho measure of the
reliability of the instructors' ratings was available.
such as that made by Tiffin (26;
011
however, studies
the reliability of merit ratings
was used as a basis for estimating the reliability of the present
ratings as 0.50)
This criterion of the validity of the Furuue mechani­
cal Assembly Test is undoubtedly affected by otner factors besides
mechanical ability, per se.
emcloyee „orth.
The retin^, is an overall estimate oi
ouch personality traits as inmustriousness, emotional
stability, ana the line, wnile important in an employee, ao not
necessarily go harm in hana ..itn expert ability on the job.
It seems
29
reasonable to state that even if tno present test were a perfect measure
of mechanical ability,
it could not give a completely satisfactory
estimate of employee worth.
(c) Machinists in a large steel mill.
The plant in which the 98 machinists, listed in Group B
above, were employed, maintains an industrial merit rating.
one of the 98 ratings ..ere available.
Ninety-
This rating is made yearly by
three superiors for every plant employee.
a previous study (26)
had revealed that the reliability of this rating is about 0.55.
The
current composite industrial merit rating was found to correlate
0.31
with Purdue mechanical assembly Test scores.
for attenuation, using
0.55
for the reliability *,f the merit ratings,
the resulting coefficient was 0.51 ± C.1G.
correlation
.hen this was corrected
Interpretation of this
coefficient as an index of validity should be similar to
the interpretation applied to the relationship between present test
scores and ratings of apprentices mentioned above.
(d) Vocational high school students.
After
8
weeks of schooling,
stuients in the vocational
high school, mentioned in Group D above, receive a rating.
This rating
was found to have a reliability of 0.64 ± U.G2 (see appendix d ) .
For
41 students of this school, correlation between this rat i n g and present
test scores was
0
.0
7
.
when this was corrected for attenuation, the
coefficient was 0.09 + 0.15*
Again, after one and one-half semesters,
the students are rated by their instructors.
For the same 41 students
the raw correlation between test scores and instructor ratings was 0.16.
After correction for attenuation, this coefficient was 0.20 + 0.14*
V/hile this second correlation coefficient is definitely nigher than the
30
first neither is statistically significant.
Two facts operated to
reduce the value of this measure of the t e s t ’s validity.
First, the
rather small group of 41 students gives quite high standard errors*
Second, the test was somewhat too difficult for this group.
There was
a high proportion of near ceiling scores (see Appendix. E.)
A brief summary of the statistical data included in this
discussion of the validity of the Purdue Mechanical Assembly Test is
presented in Table IV*
Table TV.
Correlation between Purdue Mechanical Assembly Test
Performance and Various Criteria.
Group Measured
Criterion
Raw Correlation Corrected Correlation
+ Standard error.
Purdue Univer—
Grades in all
sity Engineering courses for two
Students
years
—0.02
—0.02 + 0.09
Purdue Univer—
Grades in Engin—
sity Engineering eering drawing,
Students
Descriptive geom­
etry and shop.
0.32
0.43 ♦ 0.08
Machinists from
9. large steel
mill
Industrial merit
ratings
0.31
0.51 + 0.10
Machinists'
apprentices
from a large
machine tool
industry
Ratings of instruetors in the
trade school
0.21
0.34 1 0.13
Vocational high
school students
Instructor ratings
after first six
weeks of school
0.07
0*09 * 0.15
Instructor ratings
after one and onehalf semesters
0.16
0.20 ♦ 0.14
Vocational high
school students
31
3»
Effects of •braining*
The 4 8 apprentices in the machine tool
industry, listed in Group C.above, had been in training for various
amounts of time.
For some, this was as little as 300 hours, for others
as much as 8,000 hours (see Appendix C).
Correlation between hours of
training and Purdue Mechanical Assembly Test scores for this group of
48 was 0*43*
Wh e n corrected for the unreliability of the test, this
coefficient was 0.49 + 0 .11 ,
This relatively high degree of relationship might be accounted
for in different ways.
First, it is probable that those individuals
possessing a greater amount of mechanical ability would be retained in
the training program, whereas those with a lesser amount wou l d be
dismissed from time to time as their lack of ability became apparent
to their superiors.
If this were true, then the correlation between
training and test performance would be an indirect evaluation of the
correlation between ability and test perforiaance.
Second, it is pro—
baHLe that the training received in this particular industry's trade
school was such that those with greater training learned more of the
techniques for solving the test than did those with less training.
If this were true, then the correlation between training and test
performance would simply show to what extent the test measured the
effects of training.
(To help clarify this interpretation, a partial
correlation was made between test performance and instructor rating
when the effects of training were held constant).
correlation coefficients are as follows:
Correlation between rating and test = 0*34
Correlation between rating and training ** 0.48
Correlation between test and training = 0.49
The zero order
32
When training was held constant, the partial correlation between rating
and test performance was 0.14*
Such a reduction, from 0.34 to 0.14*
may be interpreted as indicating a relatively important training effect.
In this connection, however, a question as to the validity of the
instructor ratings may be raised.
Is the instructor rating a fair
index of the ability of the student?
Or, is the instructor rating
influenced markedly by the length of time a student has been in train­
To aid in discerning the true relationship in regard to questions
ing?
of this ty:_e, a partial correlation was made between instructor rating
and training when the influence of mechanical ability was ruled out.
In order to make such a computation it ..as necessary to inter] ret the
Purdue Mechanical Assembly Test score as a valid measure of mechanical
ability.
When mechanical ability, so measured,
was held constant, the
correlation between instructor r:ting and training was 0.38.
This is
a reduction of only ten points from a zero order correlation coeffi­
cient of 0.46.
Thus, one may reason that instructor rating is affected
considerably by training (a longer p>eriod of association), and not
so much by ability.
Either of the above interpretations may have some statistical
suoport, yet neither appeal's to be tenable on the- basis o f the limited
Further study of the effect o f training, o r e x p e r i e n c e ,
data available.
ir cn success on the i-urdue Mechanical Assembly Test is needed.
4
.
Correlations with a g e .
The 48 m achinist’s apprentices,
identified above in Group C, ranged from 18 to 27 years of age.
Correlation between age and the present test for this group yielded
a coefficient of
was
0.15
*
0
.1
2
.
0
.1
3
.
-lien corrected for attenuation,
this coefficient
’While this is not a significantly high positive
33
correlation,
training,
it may be said that, inasmuch as this group was in
the older apprentices might perform better in this test
because of more inclusive mechanical experience.
The age record of 91 of the 98 machinists,
was available.
in Group B above,
This covered a range of from 20 to 54 years.
Corre­
lation between age and scores o n the Purdue Mechanical assembly Test
for these 91 persons gave an r of —0.22.
7/hen corrected for the
unreliability of the present test, this coefficient «as
0.25
This differs significantly from the ioregoing positive value.
group was composed of trained machinists,
♦
0
.1
0
.
This
however, and an age range
up to 54 years quite likely includes some less proficient persons in
the upper brackets.
5.
Correlations with general intelligence.
Entering freshmen
at Purdue University all are measured by the American Council on
Education Psychological Examination.
Scores on this test (1939) for
119 of the 3.28 students included in group A above, correlated —0.002
+ 0.09 with the present measure.
Assuming the ACE examination to be
a valid measure of general intelligence,
it may be said that there is
no significant relationship between performance o n the Purdue Mechani­
cal Assembly Test and general intelligence for this group.
The 43 apprentices,
listed in Group C above, were given the
wonderlie Adaptation of the Otis S-A examination, known as a Personnel
Test.
Scores on this test when correlated with scores o n the Purdue
Mechanical Assembly Test gave an r of 0.33»
'/hen this was corrected
for attenuation, using the published reliabixity for the Personnel
Test, the correlation was 0.39 ± 0.12.
This significantly positive
correlation differs considerably from the r of - 0 . 0 0 2
found for the
34.
Purdue student group.
One possible explanation of the difference
might b e traceable to the fact that the two groups were selected
differently.
Engineering students are selected, in part at least, on
the basis of g e neral intelligence;
are selected,
Thus,
whereas m a c h i n i s t fs apprentices
in part at least, on the basis of mechanical ability.
it m a y b e reasoned,
for a group having a relatively high and
relatively uniform general intelligence there appears to b e little
relationship between general intelligence and performance on this
test.
However, for a group having a relatively high and relatively
u n i f orm degree of mechanical ability,
there is a significantly
positive relationship between general intelligence and success on
this test.
6.
Correlations with other t e s t s .
given to sorie of the groups listed above.
of persons involved,
Various other tests were
The group tested, the number
the tests used, and the raw and corrected
correlation coefficients are itemized in Table V.
In each case the
published reliability coefficients were used in determining the
corrected correlations.
Table V.
Performance on the Purdue Mechanical Assembly Test
correlated with other measures.
Group
Engineering
Students
Eng ineer ing
Students
Apprentice
Machinists
N
119
119
48
Raw r
Corrected r +
Standard
—
Deviation
0.37
0.49 + 0.08
Minnesota Paper
Fo r m Board
0.18
0.21
Technical Information
in Industrial
Mathematics
0.39
0.47 * 0.12
Test
Minnesota Spatial
Relations Test
0.08
35.
Table V. (Continued)
Apprentice
Machinists
46
Technical Information
in Machine Shop
0.54
0.62 + 0.10
Vocational High
School Students 41
Army Beta Examination
0.47
0.56 + 0.12
A review of the data presented in Table V reveals that Purdue
Mechanical Test performance is correlated to a lesser extent with
performance on the Minnesota Paper Form iioard Test than with any of
the others.
abilities.
It seems apparent that these two tests aauxple different
The Minnesota opatial Relations Test, whi c h requires the
subject to use apparatus and measures, to some extent, an ability to
aiscrM.iinate shapes,
correlated
0.44
with the present test.
do doubt
some of the abilities required to do well on the Purdue Mechanical
Assembly Test are sampled by the Minnesota Bpatial Relations Test.
The other three tests listed in Table V all are more highly correlated
with the present measure.
/ill three unquestionably sample some
abilities sampled by the present measure.
However, all of the corre­
lation coefficients in Table V are low enough to indicate that any of
the other tests could be included in a battery which contained the
present test without causing much duplication.
7
.
Factor Analysis of sub-test Data.
Intercorrelations of
scores made on the eight sub-tests by the 2 0 3 persons -who took Form
A first are presented in Table VI.
T a b l e VI*
Purdue Mechanical assembly Sub— test Intercorrelations.
A— 1
A— 2
A-3
A-4
B—1
B—2
A—2
.46
A-3
•A'~-
.21
A-4
.69
.49
.36
B—1
.45
.43
.26
.46
B— 2
•36
.37
.25
.
oo
G
•
OJ
H
A— 1
B-3
.33
.36
.20
.34
.32
.45
B—4
.34
.39
.15
.35
.31
.30
The size of these coefficients indicates that the various sub— tests
measure some elements in common.
The question as to what and how
many are the independent variables necessary to account for this table
of intercorrelations naturally arises.
data was made.
^ factor analysis of these
After com. ut&tions were completed, the analysis
yielded t w o factors.
a. rather complete report of this phase of the
study is included in -appendix G.
The ta le of second factor residuals,
reported in the appendix, was examined, ana after checking the a z e and
dispersion of these residuals, it was decided that the extraction of a
third factor would yield only spurious results.
Loadings for the two
factors found in each of the eight sub— tests is shown in Table VII*
Table VII.
Factor loadings before rotation.
Sub—Test
A —1
A— 2
A— 3
A— A
B— 1
B— 2
B— 3
B—4
Factor #1
Loading
.749
.645
*414
.882
.597
*715
*514
*464
Factor #2
Loading
-.121
.303 -.137 -.407
.22 $
- 1.68 .123
.185
37
Xn the light or the values in Table VII, and after an analysis
of the geometrical representation o f these values shown in Appendix. G*
and an analysis of the tests themselves, it was concluded that a 45 °
rotation of the axes would give the most meaningful interpretation of
the factors.
The factor loadings, after rotation, are shown in
tabular form in Table VIII*
Table VIII*
Interpretation of the factors, in terms
Factor loadings after rotation*
Sub-test
A—1
A—2
A-3
A-4
B-l
B-2
B-3
B-4
Factor #1
Loading
.614
.242
.390
.911
.264
.625
.277
*197
Factor #2
Loading
.443
.671
.196
.335
.5«0
.387
.451
.459
of their psychological meaning, was derived in part f r o m examination
of the tests themselves,
during examination.
and in part from comments made b y subjects
Sub— tests A— 1, a— 3, A —4, an d B—2, most heavily
loaded with Factor #1, present novel mechanical situations to a greater
extent than do those sub— tests less heavily loaded.
During experimenta­
tion these four sub—tests provoked critical and disparaging comments,
from persons who could be called mechanically sophisticated, to a
greater extent than aid the other sub— tests.
It seems logical to
identify this factor as an abi ity to deal with mechanical problem
situations which are n e w 9 a kind of mechanical adaptability, or,
perhaps, an insight into mechanical relationships.
Sub— tests A—2, B—1,
B-3, and B-4, while not consistently different from the others, appear
-fo
to involve a/^greater extent the use of more conventional mechanical
problem situations.
In working these problems, the mechanical soph­
isticate frequently expressed a satisfaction with certain of the
mechanical situations included.
These are the sub— tests most heavily
38
loaded with Factor #2., which factor was identified as an experiential
factor.
It should b e pointed out, however, that i^ither factor can b e
clearly associated with any sub— test, or combination thereof.
No
sub-test is a very pure measure of either ability, and each sub—test
appears to measure both abilities to some extent.
Additional research might reveal a somewhat different set of
factor loadings if made upon test scores derived from a population
more heterogeneous.
Research of this type is recommended.
It is
recognized that only if the findings submitted here can be substant­
iated by similar findings elsewhere, are these comments valid for
this test.
39
V
SUMMARY AND CONCLUSIONS
The Purdue Mechanical Assembly Test, wa s designed a n d constructed
to be similar in principle to the Stenquist Mechanical Assembling Test
and the Minnesota Mechanical Assembly Test.
characteristics,
It embodies certain
i.e., sturdy, precision construction, and non— stereo­
typed problem situations, which appear to be improvements upon those
tests wnich <iere its prorotypes.
test.
might sub— tests problems compose the
They are divided into two forms of four sub—tests each.
The
sub— test order of administration is, in each form, from the simple to
the complex.
Procedures for the administration of the test have been
standardized.
Three—hundred— thirty— eight persons representing five different
groups were tested.
From the scores of these persons Z— scores for all
raw scores for both forms of tne test have b e e n computed.
On the
basis of these 338 scores, the total test reliability was found to
be 0.77 + 0.02.
The method used to determine this coefficient was
that of correlating scores on the two forms of the test.
Four criteria were used to evaluate the validity of the test in
certain practical situations,
(l) Correlation ..j.th overall scholastic
achievement in college e n g i n e e r i n g
-0.02 ± 0.09 for 119 students.
to activities m o r e
Yihen scholastic achievement w; s limited
precisely mecnanical, namely, shop, o r aiming and
aescriptive geometry,
Q.M3
curricula yielded a coefficient of
coirelation ..ith the test gave a coefficient of
0.06 for the same group.
not a valid measure of success
From this it appears that the test is
in curricula loaded wi t h academic studies,
although it is helpful in estimating probable success in those courses
w hi c h are more mechanical.
(2) Correlation with industrial merit
ratings for 91 machinists in a steel mill gave a coefficient of 0.51
± 0.10.
In v i e w of the fact th; t industrial merit ratings are an
appraisal of over— all employee worth, it appears that the test is a
reasonably satisfactory instrument to include in a battery for select­
ing machinists in a steel mill.
( 3 ) Correlation with instructor rat­
ings of 48 apprentices in a machine tool industry yielded a coefficient
of 0.34 + 0.13*
While such a rating connotes much the same sort of
thing as does a merit rating, the test apparently is less valid in
this situation trian in the selection of trained machinists.
Correlation w i t h instructor ratings,
(4 )
after the first 8 weeks of school,
for 41 vocational high school students gave a coefficient of 0.09 ±
0.15•
Correlations wi t h ratings by the same instructors for th same
student group after one and one— half semesters gave a coefficient of
0.2C ± 0.15.
too difficult.
It was observed that for this group the test seemed
This observation, coupled wi t h the lack of signi­
ficant correlation with the criterion leads to the conclusion that the
test is not likely to b e especially valuable in a vocational high
school situation.
It is realized, however,
that such a small group,
41 subjects, does not present data fro;.'*, which conclusive findings
emerge.
A further study w i t h subjects of the secondary schools is
needed.
On the basis of data gathered from 48 machinists*
apprentices it
appears that training is a relatively important contributing factor
to success on the test.
“hen effects of training were held constant,
correlation between instructor rating and test scores was only 0.14*
41
It may b e reasoned, however, that the tendency for instructors to
rate apprentices with more training higher than those with less train­
ing, may reduce the validity of this finding.
There appears to be a positive correlation between test paform?ance and age for younger groups, and a negative correlation when an
age range great enough is involved.
For a group of 48, ages 18 to 27,
the coefficient was 0,15 ± 0.12; for a group of 91, ages 20 to 54,
the coefficient was — 0.25 ± 0.10.
Two correlations between test scores an d measures of general
intelligence were made.
For 119 Purdue University students the
coefficient was — .002 + 0.09 when the American Council on Education,
Psychological Examination, was used to measure general intelligence.
For 48 machinists apprentices the coefficient was 0.39 ± 0.12 when the
vJonderlic adaptation of the Otis S—A examination w a s used.
Fro m these
findings it seems that there is no significant correlation between
eneral intelligence and test scores for persons of college grade;
however,
for a group having relatively uniform mechanical ability,
there is a positive relationship.
A factor analysis of the sub—test intercorrelations revealed the
presence of two factors.
One of these was identified as mechanical
insight, the other as mechanical exjjerience.
In these calculations
scores from 203 persons selected from all of the groups tested were
used.
Vihether or not similar factor loadings and similar findings
«»ould result from such a study on a different group is a question
that can b e answered only by an additional study.
It is recognized that this study has accomplished only limited
results.
In the general area of vocational guidance,
the test may
42
pro&e to b e useful.
In the present ^udy the findings did not indicate
that the test was particularly suited to use in secondary schoolsj
however, only a limited number of secondary school students were
tested.
Further research in this area is desirable.
Likewise, the
effect of training in an industrial situation u p o n test performance
is a study not satisfactorily completed here.
Whe t h e r the test is
a fair measure of general mechanical training, or whether it can be
used to sample an individual*s capacity to receive training,
questions th t.additional research might answer.
are
Also, the nature of
the fundamental abilities samplea b y this test is not conclusive on
the basis of the findings of this study.
Additional factor analyses
on other populations may be helpful in this area.
43
BIBLIOGRAPHY
1.
Baker, H. J. and Crockett, A. C., Detroit Mechanical Aptitude
Examinations for Boys and Girls.
Company,
2.
Public Schools Publishing
1929.
Bennett, G. K., Test of Mechanical Comprehension, The Psychologi­
cal Corporation, New York, 1940.
3-
Bingham, V/. V., Aptitudes & Aptitude Testing, Harper & Brothers,
1937.
4.
Coover, S. L . , The Nature and Measurement of Mechanical A b i li t y t
Paper presented to I. I. E. A., April 1942.
5.
Cox, J. 77., Mechanical Aptitude. Methuen & Co., Ltd. 1928.
6.
Dewhurst, J. A., Personnel Selected and Trained in Milwaukee
on
Scientific Basis. Electric Railway Journal, 67 (1926, pp. 624—629).
7.
Fontegne, J., L*Orientation Professionelle et La Determination des
Aptitudes. Paris, 1921.
8.
Garrett, H. E., and Schneck, K. R., Psychological Testa Methods and
R e s u l t s . Harper and Brothers, 1933.
9.
Gerhardt, P. ?/., Scientific Selection of Empl o y e e s . Electric EailJournal, 47, 1916, pp. 943-945.
10.
Harvey, 0. L.,
’Mechanical a p t i t u d e 1 or M e c h a n i c a l A b i l i t y 1?
A Study in M e t h o d , Journal of Educational
.
Psychology, 22, 1931,
pp. 517-522.
11.
Hull, Clark, Aptitude Testing, "forld Book Company, 1928.
12.
Juhasz, A., Die 'Kuse* der Psychotechnik.
Z, Ang. Psych., 33,
1939,
pp. 456-464.
13.
Likert, R. a n d ^uashe,
1m. H., Revi s e d Minnesota Paper F o r m Board
Test. Psychological Corporation, 1939*
44
14.
Link, H. C., Employment. Psychology. MacMillan Company, 1919*
15 .
ikiac Quarrie, T. W., a Mechanical Ability Test.. Journal of Personnel
lies ear ch 5 , 1927, pp. 329-337.
16.
Mtinsterberg, H., Psychology a..d Industrial Efficiency. Cambridge,
1913.
17.
O'Connor, J., Born That Wav. New York, 1929, William & Wilkins, 1928.
18.
O'Rourke, L. J., Mechanical Apt.it.ude Tests, Psychological Corpora­
tion.
19.
Paterson, D. G., Elliott, R. M., Anderson, L. D., Toops, H. A.,
Heidbreder, E., Minnesota Mechanical Ability Tests. University of
Minnesota Press, Minneapolis, 1930.
20.
Patten,
E. F., An Experiment in Testing; Engine Lathe Aptitudes. J .
Applied Psychology,
21.
7, (1923) pp. 16-29.
Rupp, H., Untersuchung. zur Fahrerprufung ker aer Deutscher Reichs—
post und bei der stadtischen strassenbahnen du gemeinde ?7un.
Psycnot,
22.
Z, 1., 1926 pp 157-164, 199-220.
Seashore, R. H., Stanford Motor Skills U n i t , Psych. Monog., 39,
1928, pp 51-64.
23.
Shellow,
S. M., Reaearch in Selection of Motormen in Milwaukee.
J. Pers. Res. 4 (1925 pp. 222— 237.
■ 24.
Stenquist, J. L., Measurements of Mechanical Ability. Columbia
University Contributions to Education, No. 130, New York, 1923.
25.
Tiffin, J., Purdue Vocational Achievement Tests, Purdue University
-L. ■/
26.
-•
Tiffin,
J. and seashore,
S. - ’’Merit Rating” - Publication of
Carnegie— Illinois S t eel Cor}oration — 1940.
27.
Toops, H. a., Tests for Vocational Guidance of Children Thirteen
to sixteen. Teachers College, Columbia University, Contributions
to Education, I4o. 136,
28.
(1923)
Vi teles, E. E . , Industrial Psychology, pp. 229,
•*.
• Norton & co.,
Inc.
29.
Johnson, A. P., The Relationship of Test Scores to Scholastic
Achievement of 244 Engineering Frefrmen Entering: Purdue University
in September. 1931. Ph. D. Thesis, Purdue University, 1942.
T
APPENDIX A
Test Data — Purdue University Engineering Students
Key to Columns:
(1) Purdue
Mechanical
Assembly Test,
Raw Scores, *orm A*
(2) Purdue
Mechanical
Assembly Test,
Z— Scores, Form A.
(3) Purdue
Mechanical
Assembly Test,
Raw Scores, Form B.
(4) Purdue
Mechanical
Assembly Test,
Z—Scores, Form B.
(3) Scholarship Index.
(6) Minnesota Paper Form Board.
(7) Minnesota Spatial Relations.
(8) American Council on Education Psychological Exam.
(9) Shop-Drawing—Descriptive Geometry Scholarship Index.
(82 Subjects taking Form A first)
(1)
(2)
(3)
(4)
23:55
1.10
28:17
♦ 32:30
.33
Appleton
15:42
Ark in
.21
(5)
*
6
(6)
■*
10
(7)
■a
13
(8)
*
3
3.7
22:10
1.20
8
10
11
9
3.7
2.19
15:28
2.36
6
11
13
11
4.6
32s$0
0.33
27:05
.38
9
11
12
7
4
Balcom
27:55
.88
18:46
1.78
6
13
14
7
3.5
Barth
26:57
.99
14:20
2.52
11
13
16
13
4.3
Baughman
40:02
-.49
32:36
-.53
6
11
13
10
—
Bell
33:05
.27
27:00
.36
—
—
—
2.7
Bitler
41:25
0.60
30:35
0.20
5
4
8
4
Blair
35:19
.16
25:15
.71
9
13
13
12
4.3
Bodley
36:55
-.11
29:44
— .04
8
12
10
8
2.5
Abel
Allen
—
(9)
—
^ C o ^ ^ e d scores taken from research of A. P. Johnson, Purdue University.
(1)
(2)
(33
(4)
(5)
(63
(7)
(83
(9)
3.3
Bradshaw
42 55
-.77
3c-:12
-.12
6
6
12
5
Bretzlaff
38 35
-.33
22:45
1.12
7
10
12
10
Brick
45 57 -1.10
32:57
-.53
10
11
9
7
3.2
Brooks
43 20
-.82
26:05
.54
4
10
9
8
2.3
Buehler
20 35
1.64
17:47
1.94
8
7
8
11
4.0
Huehier
22 20
1.48
29:20
.05
6
10
15
11
4.0
Bus s ing
43 15
0.82
34:50
0.86
11
9
7
15
4.2
Cartiaell
15 10
2.25
15:57
2.27
13
11
14
15
6.2
Chana
37 uO
0.16
27:21
.38
7
12
11
2
—
Cherry
16 30
2.08
16:47
2.11
7
12
14
3
4.0
Christen
26 32
0.99
31:22
-.28
8
9
10
15
3.8
Chun
50 00 -1.59
44:15
-2.43
7
12
13
6
3.0
Concannon
32 31
.33
32: 5c
-.07
3
11
13
2
2.7
Cosley
31 15
.49
34:65
-.86
5
9
8
5
2.2
Cooinbes
^6 16 -1.15
30:43
11
11
6
12
4.0
DeCaiap
30 3o
.55
3o: 05
-.12
10
11
11
16
4.70
Elliott
25 16
1.15
22:15
1.20
10
10
12
5
4.0
Farrar
39 30
-.44
25:65
.62
9
7
9
5
3.7
Finkbiner
29 46
.66
17:33
1.94
6
7
5
1
3.7
Fletcher
30 55
.55
17:30
1.94
7
12
13
11
3.0
Flick
34 25
0.16
28:58
.13
4
6
9
9
3.7
Fraser
34 15
.16
29:20
.05
Gerkin
40 32
-.55
32:13
-.45
Gilbert
43 30
-.88
36:00
Harden
41 30
-.66
34:00
Heine
36 15
-.05
17:10
—
—
—
3.3
5
9
13
4
2.8
l
H
•
H
H
—
3.75
6
11
4
3
3.3
-.78
9
7
11
10
4.3
2.03
3
10
13
4
1.5
3
(2)
(3)
(4)
(6)
(7)
(8)
(9)
7
14
14
13
5.2
Herron
21:15
1.59
24:40
.79
Hineline
31:55
•44
33:05
-.61
—
—
—
—
3.5
Hines
30:22
.60
19:22
1.70
—
—
—
—
3.4
Jett
37:10
-.16
40:15
-1.77
8
10
12
5
3.5
Karn
25:35
1.10
29:17
-05
10
10
12
9
4.4
Korte
34:25
.16
16:05
2.19
9
9
11
4
5.2
Koch
46:50
-1.21
25:30
.62
9
9
6
13
4.0
Kintzing
30:12
.60
17:00
c'v
O
•
(5)
6
11
14
7
3.5
Koskinas
30:27
•60
18:02
1.86
2
12
12
1
3.0
Lee,
37:2-
-.16
28:52
.13
10
11
12
9
4.0
Lottes
46:uG
-1.15
3o:35
-.20
9
13
13
13
3.7
McDonald
37:48
-.22
31:^0
-.28
13
14
15
12
5.7
Miyamasu
49:20
1
•
•b
00
(1)
29:00
.05
8
13
13
7
4.5
Nelson
33:40
.22
26:25
.54
11
14
16
12
—
Nelson
39:55
— .44
32:45
-.53
9
8
11
14
3.5
Neyhart
31:25
.49
33:00
-.61
9
6
8
4
3.8
0 *Grady
40:45
-.55
24:35
.79
10
10
14
14
4.0
Oliver
49:10
-1.48
34:04
-.78
6
6
1
5
3.2
Osborne
39:30
— .44
28:55
.13
10
8
9
16
3.8
Owens
50:00
-1.59
33:02
-.61
8
14
9
13
3.8
Park
26:10
1.04
15:40
2.27
8
15
14
13
4.7
Finkham
30:55
.55
29:30
-.04
12
9
10
14
5.0
Paxenberg
23:00
1.37
21:40
1.28
6
9
8
15
—
Guakenbush
37:02
-.16
20:15
1.53
7
8
12
3
3.7
Rogers
27:52
.88
26:15
.54
4
10
7
5
2.7
Rose
44:23
-.93
31:30
-.37
.
—
—
—
—
1.5
(2,
(3)
(4)
(5)
(6)
(7)
(8)
(9)
Ross
21:35
1.54
20:10
1.53
12
13
12
12
5.2
Ruggles
44:30
-.99
30:15
-.12
10
9
9
15
4.0
Sage
41:35
— .66
39:50
-1.69
6
11
11
5
3.0
Schoonover
38:55
-.33
31:10
-.28
6
6
7
2
3.7
Shoaf
47:55
-1.32
33:05
-.61
4
11
2
2
2.7
Siff
29:10
.71
21:05
1.37
9
6
6
8
4.5
Steele
45:13
-1.04
41:45
-2.02
7
10
11
7
4.0
Streed
35:15
.05
34:55
-.86
14
11
9
13
6.5
Streeter
43:10
-.£2
32:20
-.45
8
8
10
14
3.0
Thran
32:20
.38
18:00
1.86
8
7
10
14
4.0
Ullom
32:55
.33
27:42
.29
8
8
11
10
3.7
Von Behren
40:00
-.49
28: wG
.21
—
—
5.7
V.'arren
40:30
-.55
31:55
-.37
10
5
8
12
2.4
Weigel
35:50
0.0
18:00
1.86
11
15
10
13
4.3
45:38
1
H
•
<D
(1)
20:00
1:53
8
10
13
3
3.6
Welsh
50:00
-1.59
36:05
-1.11
9
9
9
13
4.0
Westfall
42:50
-.77
35:40
-1.03
9
12
13
9
4.0
’Vinner
47:20
-1.26
30:55
-.20
8
11
13
7
4.0
’"/hippo
39:55
— .44
21:45
1.28
12
16
13
14
5.0
/einfurtner
—
—
(46 Subjects taking Form B First)
Beaser
35:40
-.27
40:35
-.85
9
9
12
10
2.2
Best
32:25
.01
23:00
1.31
11
13
13
7
4.7
Brown
34:43
-.19
31:10
.32
9
12
10
15
Chang
45:35
—1.06
50:00
-2.01
6
9
7
0
2.5
Cooke
46:34
-1.14
38:25
-.54
5
8
6
13
2.0
Curry
42:00
0
.
1
37:05
-0.42
13
11
9
15
4.3
—
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
Duncan
25:20
.56
30:45
0.38
9
10
14
15
3.7
Eichenberger
19:57
1.00
19:30
1.74
9
11
14
7
5.0
Emhoff
41:55
-.77
44:40
-1.34
9
8
13
4
4.0
Fabiani
23:10
.72
31:33
.26
6
11
13
5
3.5
Feik
40:25
-.63
44:30
-1.34
13
11
10
9
4.0
Frantz
17:40
1.16
31:10
0.32
14
7
11
15
5.1
Gamble
24:30
.60
39:55
-.72
9
12
12
9
4.0
Hadley
24:02
.64
24:36
1.12
13
13
12
16
4.8
Harper
40:53
-.67
36:41
-.36
8
12
12
12
4. 6
Hebenstreit
40:05
-.63
36:30
-.36
10
e
8
9
4.3
Henderson
27:35
.37
34:45
-.11
5
9
2
4.0
Hess
47:25
-1.18
39:15
-.66
7
7
10
14
3.0
Hoban
23:15
.72
21:30
1.49
5
11
12
9
3.8
Jones
35:05
-.23
28:25
.69
7
10
7
14
4.0
Kieffer
33:57
-.11
33:40
0.01
7
9
14
12
3.0
Landau
45:00
-1.02
25:20
1.06
8
9
8
14
3.5
Leifheit
49:05
-1.34
43:05
-1.15
9
9
10
4
3.8
Leighton
23:30
.68
32:30
.14
9
10
11
8
3.6
Lewiecki
32:45
-.03
19:15
1.80
12
10
11
13
4.8
Love
38:35
-.51
39:45
-.72
12
13
6
15
4.7
McAllister
26:11
.45
21:20
1.55
10
9
11
16
5.0
McCurdy
34:50
-.19
31:45
.26
—
—
—
—
2.3
Magner
12:25
1.59
11:05
2.78
—
—
—
—
6.5
Morehouse
27:45
.37
14:15
2.41
lu
16
13
11
3.0
Morgan
18:36
1.08
19:05
1.80
8
10
10
11
4.2
n
(3)
Neuman
33:25
-.07
33:50
Olesen
25:23
.56
Orr
36:45
0 ’Shaughnessy
35:30
Pennington
42:15
Purdy
(5)
(6)
(7)
(8)
(9)
.01
9
13
13
8
3.7
33:30
.01
7
12
13
7
2.8
I
(2)
32:45
.14
11
12
7
14
4.0
1
»
IV>
(1)
»
UJ
Vn
6
40:40
-.85
10
9
9
10
3.7
-.79
32:21
0.20
5
lo
10
3
2.8
39:51
-.59
28:05
.69
lo
12
8
8
—
Reinhardt
24:50
.60
29:45
.63
3
13
14
3
3.5
Sauter
37:10
-.39
35:50
-.23
—
—
—
1.0
Sheehy
32:07
.01
23:30
1.24
6
13
14
14
3.8
Swartz
34:17
-.15
25:15
1.06
2
7
7
3
2.4
','eaver
35:49
-.27
31:10
.32
10
7
10
9
4.0
Vebster
44:01
-.95
32:00
.20
6
12
11
11
4.0
.Jewee
25:15
.56
23:40
1.24
5
11
13
2
3.3
Zeigler
30:45
.13
41:50
.97
11
11
11
14
4.5
Zimmerman
24:00
•6 4
24:30
1.12
9
6
11
15
3.7
(4)
—
7
APPENDIX B
Test. Data — Carnegie— Illinois Steel Company, Sheet and Tin Lill,
Gary,
Indiana, Machinists.
Key to Columns:
(1)
Puraue Mechanical ^.ssejnbly
Test, Ra»,r Scores, Form A.
(2)
Purdue Mechanical Assembly
Test, Z— Scores, Form A.
(3)
Purdue Mechanical assembly
Test, Raw Scores, Form B.
(4)
Purdue Mechanical assembly
Test,
Z—Scores, Form B.
(5) Age.
(6) Industrial Merit Rating.
(57 Subjects taking Form A first.)
(1)
(2)
(3;
(4)
(5)
(6)
Bachman
25:16
1.15
20:49
1.45
45
350
Eianco
46:46
-1.21
37:24
-1.27
48
349
Elohra
31:45
.44
23:24
1.04
—
339
Boutilier
32:00
.38
15:32
2.27
31
361
Boyer
82:40
.33
33:51
-.70
4o
351
Carpenter
28:47
.82
34:13
-.76
49
314
Chianamonti
35:33
0.00
32:40
-.53
43
360
Cice
41:18
— . 60
22:46
1.12
24
—
Corwin
30:2.1
.60
3o: 20
-l.il
23
—
Cr is.ian
45:31
-l.lu
39:45
-1.69
21
—
Uobry
22: 33
1.43
22: 54
1.12
31
363
Mrckson
37:13
— .1 6
28:28
.21
12
325
Fickes
38:06
-.27
35:09
-.94
21
303
Finney
36:46
-.33
33:13
-.61
25
364
Foerder
29:37
.66
31:30
-.37
55
—
8
(1 )
Gniemfash
32:54
Graham
14:27
hlehenic
(2 )
(3)
(4)
(5)
(6 )
24:45
.79
28
—
2.36
16:53
1.78
25
391
41:36
— .66
34:03
-.78
22
—
Hocza
48:59
-1.43
42:38
-2.18
47
350
Hoov.er
34:43
.11
36:34
-1 . 5 2
27
399
Johnson
47:15
-1 . 2 6
26:00
.54
37
—
Jurov
37:33
-.22
46:54
-2.64
45
325
Kiraly
43:39
— .66
28:33
.13
38
364
Kirche
22:29
1.46
29:18
.05
42
350
La ins
34:54
1.10
22:55
1.12
—
Lamb
29:06
.71
22:55
2.60
36
364
Leyton
37:01
-.16
46:01
-2.76
24
—
I.Iark
50:00
-1.59
36:53
-1.19
50
325
L'itchell
47:02
—1.26
42:17
-2.10
54
345
’'itchell
21:56
1.54
26:16
.54
39
373
? orris
30:03
.60
18:54
1.78
36
373
Lyer
41:32
-.66
30:46
-.20
51
—
Oaell
IB: 37
1.86
31:23
-.28
41
352
Pavlinski
31:22
.49
29:24
.05
22
—
Price
31:53
.a*6
33:47
-.70
46
342
Friddy
25:29
1.15
38:39
-1.52
46
358
Rend
29:33
•66
29:11
.05
35
—
Robbins
34:02
.16
17:37
1.94
22
Seashore
3,: 11
.27
24:52
.79
—
—
Shabaz
23:32
1.32
28:25
.21
21
—
Shepperd
38:12
-.27
36:51
-1.19
23
—
Shimkus
50:00
39:30
-1.69
26
237
Shuman
38:05
40:54
-1.85
25
367
.33
-1.59
-.33
—
(2)
(4)
(5)
(6)
31:13
.49
34:39
-.86
48
369
Snyder
31:32
•44
36:39
-1.19
30
—
Spangle
41:10
— .60
48:04
-3.09
39
252
Steele
36:07
•1
o
(3)
S m it h
32:00
-.45
26
404
Stone
34:32
.11
36:25
-1.11
32
340
Sundquist
29:16
.71
25:49
.62
43
325
Tison
30:04
.55
41: o2
-1.93
21
—
Vaidik
34:30
.16
19:18
1.70
26
322
Valentine
36:54
-.11
26:22
.13
36
—
T argo
49:08
-1.48
41:19
-1.93
52
357
27:42
0 .8 8
36:49
-1.19
21
291
/atkins
..att
30:02
.66
34:36
-.86
36
345
Yeager
42:13
:>
•
H
38:57
-1.52
51
—
illis
16:24
1.92
16:20
1.86
22
l
(41 2iubjects ta king Form B first.)
39
362
—
345
-.11
52
357
H
•
1
i
I
(1)
39:31
-.72
35
352
-.19
48:05
-1.77
—
325
31:35
.05
28:57
.63
25
—
Cox
46:13
-1.10
37:42
-.48
26
279
Dond&nville
38:46
-.51
38:53
-.60
27
386
Durkovich
26:09
.49
29:36
.50
28
354
Klvin
34:47
-.18
34:38
-.11
43
346
Athey
41:25
-.71
38:20
-.54
Bailey
31:67
.09
31:23
.32
Biddle
39:16
-.55
34:48
Bullman
41:04
Calhoun
34:51
Christy
(1)
(2)
(3)
(4)
(5)
(6)
Fainley
28:34
.29
84:14
-.05
50
390
Gibson
43:44
-.91
45:19
-1.40
34
—
Glassley
37:22
-.39
37:24
— .48
51
369
Guthrie
38:44
-.51
44:39
-1.34
44
343
Hamilton
20:19
.96
30:46
.38
38
363
Hines
18:35
1.08
30:13
.44
22
—
Hogan
27:43
.37
28:12
.69
42
344
Horvath
32:44
-.03
34:38
-.11
36
349
Johnson
37:24
-.39
41:45
-.97
36
354
Kadelak
31:02
.09
42:02
-1.03
25
348
Koucier
46:27
-1.10
36:16
-.54
50
346
Labadie
33:01
-.07
33:16
.07
28
354
Law
28:43
.29
42:27
-1.09
22
—
Leets
35:00
-.23
35:56
-.23
54
343
Lenon
46:06
-1.10
49:48
-1.95
—
323
Micknik
35:41
-.27
34:30
-.11
39
386
Owen
38:27
-.47
27:09
.81
24
339
Pickering
31:33
.09
30:37
.38
46
287
Pitchford
43:07
-.8?
46:23
- 1.52
42
400
Polateniez
47:08
-1.18
42:01
-1.03
54
300
Primer
19:20
1.04
25:48
1.00
36
360
Robb
38:25
-.47
38:27
-.54
35
388
Ross
31:12
.09
37:23
-.42
35
372
Sheppard
27:05
.37
38:04
-.54
21
—
Smith
49:26
-1.34
50:00
-2.01
43
294
11
(1)
(2)
(3)
(4)
(5)
(6)
Soderstrom
31:03
.09
40:48
-.85
51
362
Springinan
47:36
-1.22
43:54
-1.22
40
314
Tison
44:07
-.95
33:30
H
O§
23
—
Yaros
41:04
-.71
40:57
.85
29
326
Zierk
26:35
.45
37:12
-.42
46
350
12
APPENDIX C
Test Data — Cincinnati Milling Machine Company Machinist*s Apprentices
Key to Columns:
(1)
Purdue Mechanical Assembly
Test, Raw Scores, Form A.
(2)
Purdue Mechanical Assembly Test, Z—Scores, Form A.
(3)
Purdue Mechanical Assembly Test, Raw Scores, Fo rm B.
(4)
Purdue Mechanical Assembly
Test, Z—Scores, Form B.
(5) Technical Information in Machine Shop,
Purdue Vocational Achievement Tests.
(6) Technical Information in Industrial Mathematics,
Purdue .Vocational Achievement Tests
(7) Personnel Test, Form D. (Wonaerlic Adaptation of O t i s ’s
5— A Higher Examination)
(£) Training Time in Hours.
(9) Ratings — Composite of Six Instructors in the Trade School.
(24 Subjects taking Form •«. first)
-
(1)
(2)
(3)
(4)
(5) (6)
(7)
(8)
(9)
778
D
Bailey
49:40
-1.54
33:04
-.61
36
20
29
Bissell
16:57
2.08
14:25
2.52 106
63
39 8811
A—
Buckles
34:40
.11
21:53
1.28 106
44
29 8332
C+
Deer
30:16
.60
13:34
2.60
90
49
28 7427
A
Dirr
16:55
2.08
15:57
2.27
48
49
19 2694
C
Dugle
22:27
1.48
16:45
2.11 106
90
45 4120
A-
Griffith
44:52
-.99
30:39
-.20
50
23
19 1261
D
Grindrof
42:24
-.71
33:33
-.70
65
40
17 1194
C
Huntington
27:50
.88
11: 52
2.94
82
71
32 2280
D '
Jacob
24:12
1.26
32:31
-.53
65
34
21 1467
X
'V.
13
(1)
(2 )
(3 )
Johnson
44:58
-.9 9
23 :22
Kapfer
2 9 :52
Loeffler
Noel
(5)
(6 )
(7)
(8 )
(9)
1 .0 4
63
39
21
7788
C-
.6 6 4 0 :2 0
-1 .7 7
52
39
22
980
c
33:00
.2 7 3 1:52
-.3 7
60
19
23
8163
c
47:50
- 1 .3 2 28:30
.1 3
66
35
19
2933
c
.8 7
90
53
28
4843
c
38:40
-1 .5 2
39
31
22
788
.1 6 30:34
-.2 0
59
32
23
2428
c
33:05
-.6 1
67
57
23
3101
c
1 .7 0 16:34
2 .1 1
70
48
34
376
c
(4)
«
Osmond
16:24
Ransdell
45:24
Ripley
3 4:25
Rolfes
3 9:55
Runyan
20:10
Stauverman
40': 4u
-.5 5
3 2:55
- . 53
71
42
21
3845
c
Snyaer
2 1 :40
1 .5 4 2 5 :5 3
.62
91
46
29
8154
B
Stauffer
32:40
.3 3 36:30
1 .1 9
84
68
32
3420
D
Talbot
2 4 : 55
1 .2 1 2 0 :5 6
1 .4 5
63
SO
39
1140
D
;eh:nan
34:29
.1 6 3 5:10
-.9 4
39
23
16
1280
B-
2 .1 4 24:05
-1 .0 4
.
-.4 4
'
c+
(24 Subjects taking Form B first)
^.bbott
24:21
.6 4 3o:45
.6 4
56
46
27
6902
/" i
U
Birms
1 1 :03
1 .6 ? 1 5:34
2 .2 3
107
83
40
4560
C
Bamkamp
2 8 :10
.3 3 2 8:4 5
.6 3
87
52
24
3936
B-
Buckles
4 m : c0
-.9 5
39:43
-.7 2
44
14
12
6330
D
Crawtis
m3:
-.1 1
39:37
-.4 8
81
74
43
946
B-
Dell
4 8 :40
—1 .30 47:00
-1 .6 5
59
63
30
1443
B+
Bberhardt
12:53
17:07
2 .0 5
98
36
22
7957
B
Ivans
2 1:19
.8 8 24:56
1 .1 2
95
77
36
8997
V-
Geier
33:55
-.11 36:25
- .3 6
73
70
37
2/4-0
Glindmeyer
31:42
.0 5 26:19
.9 4
100
65
31
8767
50
1 .5 5
c-
(1 )
(2)
(3)
(4)
(5)
(6 )
(7)
(8)
(9)
Greeley
38:13
-.47
33:50
.0 1
91
92
27
80 50
B
Havlin
34:26
-.15
2 9 :03
.57
80
32
25
954
C
Hav1 ish
18:09
1.12
31:15
.32
92
49
26
8950
A—
Ref i'ner
38:27
-.47
34:40
-.11
50
37
31
1009
D
LoPiccclo
3 6 :06
-.31
33:15
.07
65
83
37
1140
Tj
...cLo,
37:20
.39
38:41
— .60
49
49
23
1997
0+
opr* is
49: 3-
-1.38
501:00
—2 .ol
35
26
22
540
D-
li-locn
14: o7
1.43
17:90
1.98
67
35
3420
D-
Pie] PlO
27:01
•41
29:10
.37
81
85
25
90-i<9
0
^ ci: lalfuss
27:27
.41
39:25
73
59
27
2613
r-*
•
>
waller
31:15
.09
7?: 45
.75
52
67
«^9
1156
3
unj'der
O >4•j
£^
C
cO#
•^ 9
29
•
7^
4-6
31
8138
15: 54
1.32
2. 6e
94
67
21
9 I9 2
•TS.""
29: 28
.21
.26
92
48
26
yO'Ul
c
.t
3 1 :45
.
15
APPENDIX D
Test Data — Evansville,
Indiana, Mechanic Arts High School Students
Key to Columns:
(1)
Purdue
Mechanical Assembly Test, Raw Scores, Form A.
(2)
Purdue
Mechanical Assembly Test, Z-Scores, F o r m A*
(3)
Purdue
Mechanical Assembly Test, Raw Scores, Form B.
(4)
Purdue
Mechanical Assembly Test, Z—Scores, For m B.
(5) Ratings by Instructors alter first eight weeks.
(6) Ratings b y Instructors after one and one—half semesters.
(7) Arm y Beta Examination Scores.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(26 Subj ects taking Form A first)
Axton
50:00
-1.59
38:58
-1.52
38
43
81
Bray
36:21
-.27
40:00
-1.77
52
50
111
Geisler
46:34
-1.21
42:20
-2.10
42
38
76
Griswold
37:04
-.16
25:58
.62
49
37
101
Hanebutt
26:22
1.04
31:38
-.37
52
51
83
Kitzinger
46:58
-1.21
47:26
-2.92
47
48
90
Gal in
41:46
— .66
39:15
-1.60
52
51
86
Miegl
23:16
1.37
32:54
-.53
60
66
107
Morris
45:20
-1.04
29:53
-.04
49
54
92
Morris
36:07
-.05
33:16
-.61
49
56
102
Oertwig
50:00
-1.59
35:28
-.94
56
55
80
Oldham
45:40
-1.10
44:55
-2.51
50
48
79
Prince
50:00
-1.59
33:36
-.70
53
51
90
Ringham
44:47
-.99
40:12
-1.77
54
38
—
Riley
45:32
-1.10
36:00
-1.11
57
5A
116
/ _»
(J V
-y
r'
C0 J
f r-t 1
* ;y
— .£2
29:21
.05
49
45
105
ucr'.uutz
a?’:e2
—1 .2 6
‘-V:^5
.05
49
49
103
ochre) eder
46:2^
-1.57
31:22
-.28
54
61
102
-pradley
OV :42
—1. >4
30: C 5
-.12
50
46
101
Ve a^hu
5a :00
-1.5 V
41: 52
-2.U2
57
54
93
ri
U)
5o :0C
-1.59
34:39
— .86
51
54
105
M l 2da , as
34:0 3
.16
29:43
— .CA
4£
52
105
3>0 •O v.■
-1.5V
44:2o
49
52
86
*0<3• <1^
.82
31:33
— •—
r 9i
i
40
48
£9
“1
" /.
—J
-. £,
41: 51
-2.18
54
54
114
— . O _■
37 : 55
-1.36
56
oe
91
*>
■o:22
illiajas
..'elf
.o 1 1 ey
1 'J'J :~ •
50 :
42 : 3?
(17 S u b j e c t s
taking For m B first)
41:45
-.75
53:31
rr
— •3 3
46
54
IOI
/ '7 • \;-
-1.11
V>I c
^ •*o.3
5c
rc
104
.r l a y e r a
22 :25
.c?
'. n • ~ _
>
* • —■/
—
.42
5o
65
115
Gfcn--<er
32:41
— • a
A ~ IA ^
<
:1 1
—„'
3c
75
J ill:er t
5a :v.
“-i.. t--.
:■-•^ -
—a-.W —
-
—
—
J-: _ 1 .e.
a a :.>-+
-.91
3 'i
-A J
-.72
49
49
£5
.o.i !4
31: 5 a
-.27
i3•
.•{
->
•.^
-1.2?
A9
36
93
. ail
3 7 :-a
—
.01
A6
58
la 2
dliver
2 1 : 54
.21
'/ • 1 + 3
. i:
A5
41
101
laic;::
a^
'< • • »
—
*— a
.62
"■ P . i A
^
•C- • j_'
r
.2^
55
109
2chler
3 9 :2 v
alas
*4.--iJJ'lt-"tfc:
_.ry ant
'lam.
—i 1 m ^
i
.39
3
3•
^
vr
-.54
r>r
it2 : 57
— 1 .-V
ri
60
101
50:17
.17
2 9 : 2a*
.57
.'.8
49
107
30:11
-.47
36:^3
-.29
c
59
116
_
17
(1 )
■■ i
16:10
" it
-
f
V.oV
1.10
01 :0 5
-.91
01:12
-.91
-.11
3 0 :o 6
.L 0
.01
37:10
-.08
L~
33:51
■_lo on
,'cie:
»r
.
10
(7)
51
02
c5
6
62
62
/•/
oo
Ar.
99
'.
< :-.re ;:eaic-,ns o: io-
e:;r;s 7.re
O.t- l i c O
(6 )
U)
^ J
L i l t ;
c ;
r t t i u
3 I n
C . 62 .
+
r\c
or :;ore
o . o 2 0 .
- e s ;
18
APPENDIX E
Test, Data — Unclassified Adults
Purdue Mechanical Assembly Test Time
Subjects taking Form A first
Form A
Raw Scores
Form B
Z—Scores
Raw Scores
Z—Scores
Blackburn
25:30
1.10
21:30
1.28
Exton
33:35
.22
23:45
.95
L'arkaw
34:02
.16
29:35
.13
Price
25:36
1.10
15:25
2.36
Rhodes
37:55
-.22
22:50
1.12
Skallerup
32:56
.33
14:10
2.52
Smith
26:27
1.04
16:12
2.19
Thompson
23:00
1.37
14:40
2.44
Trimton
35:51
32:37
-.53
•Yagar
30:55
-55
27:58
.29
Ydowka
23:49
1.32
17:25
2.03
2:43
1.43
28:30
.13
Connor
48:10
-1.37
30:45
-.20
Herrick
25:10
1.15
16:05
2.19
Moutaux
'
0.0
Subjects taking Form B first
Daugherty
24:32
.60
28:50
0.63
Green
36:00
-.35
31:32
0.26
McClure
25:35
.53
16:12
2.17
Montgomery
35:05
-.23
25:45
1.00
19
Form A
Raw Scores
Form B
Z—Scores
Raw Scores
Z-Scores
Polak
13:50
1.47
25:15
1.06
Itis ing
18:40
1.08
27:13
0.81
Rose
23:47
.68
31:32
0.26
','Jerner
33:55
-.11
26:50
0.87
20
APPENDIX F
Distribution of Sub-Test Scores.
(Form A first, N « 203; Form B first, N * 135.)
Class Interval
Form A
Form B
Class Interval
Form A
Form B
In Minutes
First,
First,
In Minutes
First,
First,
Number
Number
Number
Number
(Sub— test A— 1)
136
50
2.9
2
3
4.90-4.99
3
2
2.8
3
3
4*3
2
2
2.7
0
4
4.7
3
2
2.6
2
2
4.6
3
1
2.5
5
6
4.5
3
4
2.4
0
1
4.4
2
3
2.3
0
0
4.3
5
2
2.2
2
3
4.2
3
2
2.1
1
2
4.1
4
1
2.0
3
3
4.0
6
5
1.9
0
1
3.9
3
2
1.8
3
2
3.8
1
2
1.7
1
1
3.7
0
2
1.6
1
1
3.6
1
2
1.5
1
0
3.5
5
3
1.4
0
0
3.4
3
3
1.3
0
0
3.3
3
4
1.2
1
0
3.2
6
2
1.1
0
2
3.1
5
3
1.0
0
0
3.0
1
3
0.9
0
1
5-0
V
21
Class Interval
Form A
Form B
Class Interval
For m A
Form
In Minutes
First,
First,
In Minutes
First,
First
Number
Number
Number
Numbe
(Bub— test A—2)
10.0
36
14 '
5.6
9
4
3
1
5.4
6
11
9.6
2
0
5.2
6
1
9.4
2
2
5.0
8
3
9.2
1
3
4.8
7
3
9.0
4
1
4.6
8
1
8.8
1
2
4.4
5
3
8.6
1
3
4.2
5
5
8.4
3
1
4.0
5
8
8.2
5
2
3.6
1
4
8.0
3
1
3.6
7
5
7.6
1
2
3.4
3
8
7.6
4
3
3.2
9
2
7.4
9
2
3.0
4
8
7.2
9
3
2.8
2
1
7.0
4
5
2.6
1
2
6.6
4
1
2.4
2
O
6.6
4
1
2.2
O
1
6.4
4
6
2.0
0
0
6.2
6
3
6.0
4
5
5.8
2
3
9.6-9.99
Class Interval
Form A
Form B
Class Interval
Fo rm A
Form 1
In Minutes
First,
First,
In Minutes
First,
First
Number
Number
Number
Numbe:
v
(Sub— test A— 3)
64
35
10.6
1
1
14.6-14.99
1
2
10.4
2
1
14.6
2
0
10.2
4
1
14.4
1
1
10.0
3
1
14.2
0
2
9.6
1
1
14.0
0
0
9.6
2
3
13.8
1
1
9.4
1
1
13.6
2
1
9.2
1
0
13.4
3
1
9.0
1
6
13.2
2
0
8.8
2
1
13.0
1
0
8.6
3
12.8
1
0
8.4
4
1
12.6
0
1
8.2
4
1
12.4
2
1
6.0
4
1
12.2
1
1
7.8
2
3
12.0
3
4
7.6
1
1
11.8
0
1
7.4
4
3
11.6
0
0
7.2
3
2
11.4
2
1
7.0
5
3
11.2
0
1
6.8
5
1
11.0
0
2
6.6
4
2
10.8
-1
0
6.4
7
1
15.0
•
.
0
23
Class Interval
Form A
Form B
Class Interval
Form A
Form
In Minutes
First,
First,
In Minutes
First,
First
Number
Number
Number
Numbe
6.2
2
1
3.4
2
3
6.0
1
5
3.2
0
3
5.8
2
0
3.0
3
0
5.6
1
0
2.8
3
1
5.4
2
4
2.6
2
0
5.2
4
0
2.4
0
2
5.0
3
1
2.2
2
3
4.8
2
1
2.0
5
3
4.6
0
2
1.8
0
0
4.4
6
3
1.6
0
2
4.2
2
1
1.4
0
' 1
4.0
4
1
1.2
0
3
3.8
3
1
1.0
0
2
3.6
3
2
(Sub— test A—4)
53
29
16.0
3
3
19.50-19.99
5
5
15-5
5
2
19.0
5
2
15.0
7
4
18.5
2
2
14.5
7
6
18.0
6
4
14.0
9
2
17.5
2
1
13.5
3
4
17.0
6
7
13.0
5
4
16.5
1
2
12.5
7
1
20.0
24
Class Interval
Form A
Form B
Class Interval
Form A
Form B
In Minutes
First,
First,
In Minutes
First,
First,
Number
Number
Number
Number
12.0
10
6
8.0
4
3
11.5
12
5
7.5
1
3
11.0
6
2
7.0
0
3
10.5
4
6
6.5
0
0
10.0
4
6
6.0
3
4
9.5
8
7
5.5
3
1
9.0
12
6
5.0
3
1
8.5
6
3
4.5
0
1
(Sub— test B—1)
5.0
37
37
3.6
2
2
4
0
3.5
3
0
4.8
1
4
3.4
o
2
4.7
1
0
3.3
0
1
4.6
1
2
3.2
2
2
4.5
3
2
3.1
3
4
4.4
1
0
3.0
5
5
4.3
1
1
2.9
3
1
4.2
1
2
2.8
2
3
4.1
2
1
2.7
5
2
4.0
2
2
2.6
4
2
3.9
3
0
2.5
6
5
3.8
2
1
2.4
4
2
3.7
2
3
2.3
3
4
4.9-4.99
25
Class Interval
Form A
Form B
Class Interval
Form A
Form ]
In Minutes
First,
First,
In Minutes
First,
First
Number
Number
Number
Numbei
2.2
5
3
1.3
7
0
2.1
9
4
1.2
11
1
2.0
13
7
1.1
5
3
1.9
7
5
1.0
8
2
1.8
7
3
0.9
1
3
1.7
7
6
0.8
2
0
1.6
8
2
0.7
1
1
1.5
3
5
0.6
1
0
1.4
5
0
(Sub—test B—2)
31
43
7.2
2
2
9.6-9.99
3
0
7.0
1
1
9.6
3
0
6.8
5
0
9.4
1
1
6.6
5
2
9.2
2
1
6.4
2
4
9.0
4
1
6.2
1
3
8.8
1
0
6.0
3
3
8.6
0
1
5.8
6
3
8.4
1
3
5.6
4
1
8.2
2
1
5.4
2
2
8.0
2
0
5.2
5
1
7.8
1
0
5.0
4
1
7.6
1
3
4.8
3
2
7.4
3
0
4.6
6
2
10.00
'I
26
Class Interval
Form A
Form B
Class Interval
Fo rm A
Fo rm
In Minutes
First,
First,
In Minutes
First,
First
Number
Number
Number
Numbe
4.4
8
1
2.8
8
3
4.2
6
7
2.6
6
4
4.0
7
6
2.4
7
4
3.8
5
4
2.2
3
6
3.6
7
2
2.0
4
2
3.4
9
6
1.8
2
1
3.2
9
3
1.6
1
1
3.0
16
4
1.4
1
0
(Sub— test B-3)
4
7
12.0
0
0
0
0
11.8
0
0
14.6
1
G
11.6
0
1
14.4
0
0
11.4
0
2
14.2
0
0
11.2
0
2
14.0
0
1
11.0
1
0
13.8
c
0
10.8
0
1
13.6
0
G
10.6
0
0
13.4
0
0
10.4
0
0
13.2
0
1
10.2
0
1
13.0
0
1
10.0
1
0
12.8
G
G
9.8
1
0
12.6
G
0
9.6
3
4
12.4
1
G
9.4
1
2
12.2
0
0
9.2
2
0
15.00
14.8-14.99
_ ^
27
Class Interval
Form A
Form B
Class
Interval
Form A
Form B
First,
First,
Number
Number
I
In Minutes
First,
First,
Number
Number
In Minutes
9.0
2
2
5.4
4
6
8.8
1
3
5.2
7
1
8.6
2
0
5.0
7
4
8.4
1
0
4.8
12
5
8.2
2
0
4.6
5
6
8.0
0
4
4.4
5
9
7.8
1
1
4.2
9
5
7.6
3
>2
4.0
13
9
7.4
1
2
3.8
6
7
7.2
2
1
3.6
15
6
7.0
2
5
3.4
16
6
6.8
5
1
3.2
16
2
6.6
3
2
3.0
11
4
6.4
3
3
2.8
6
1
6.2
2
2
2.6
4
2
6.0
2
2
2.4
5
0
5.8
3
2
2.2
6
1
5.6
2
4
2.0
4
2
(bub— test B-4)
94
86
17.5
1
5
2
1
17.0
2
1
19.0
3
3
16.5
3
2
18.5
1
1
16.0
4
4
18.0
5
4
15.5
5
0
20.00
19.5-19.99
28
Class
Interval
In Min u t e s
Form A
Form B
Class
First,
First,
In Minutes
Number
Number
Interval
Form A
Form B
First,
First,
Number
Number
15.0
5
2
9.5
5
1
14.5
6
3
9.0
3
1
14.0
3
2
8^5
4
0
13.5
3
1
8.0
4
1
13.0
3
2
7.5!
4
1
12.5
5
1
7.0
5
1
12.0
3
2
6.5
2
2
11.5
3
2
6.0
3
0
11.0
4
1
5.5
3
1
10.5
4
2
5.0
2
1
10.0
7
1
4 .5
0
0
4.0
1
0
29
APPENDIX G
Sub— test Factor Analysis Data.
Number of subjects = 203.
Number of sub—test =
8.
Original Correlation Matrix;
A— 1
A —2
A-3
A-A
B— 1
B—2
B-3
B—A
A—1
^62
.A6
•AO
.69
.A5
.36
.33
-3A
A—2
.A6
.21
•A9
.A3
.37
.36
.39
A-3
.AO
.21
^ 0
.36
.26
.25
.20
.15
A—A
.69
.A9
.36
.80
.A6
.80
•3A
.35
E—1
•A5
.A3
.26
.A6
^^6
.31
.32
.31
B—2
.36
.37
.25
.80
.31
.80
•A5
.30
B-3
.33
.36
.20
•3A
.32
•A5
B-A
•3A
•39
.15
.35
.31
.30
.19
'
.19
First factor loadings: and c oriLiunalit ie s
A— 1
A—2
A-3
A-A
B—1
E—2
B-3
B-A
Loading
.7A2
.638
.AA5
.856
.598
.726
.526
.As:
C orr.iunal i ty
.5506
.A070
.1980
.7327
.3576
.5271
.2767
.2 3 :
30
First Factor Residuals
A—1
—
A—2
+
A-3
—
A—4
B —1
+
B—2
—
—
B-3
+
B-4
+
A—1
+. 17?
+ .013
+ .070
+ .055
-.006
-.179
+ .060
+ .018
A—2
+ .013
+ .074
+ .056
+.048
+ .093
+ .024
+ .082
A-3
+ .070
+.074
-.021
+ .006
-.073
+ .034
+ .065
A—4
+ .053
+.056
-.021
+ .178
+ .052
+ .178
+ .110
+ .063
B— 1
-.006
+ .048
+ .006
+ .052
+ .124
+ .124
+ .005
+ .021
B—2
-.179
+ .o931
-.073
+ .178
+ .124
+ -17?
-.068
+ .051
B-3
+ .060
+ .024
+ .u 3a
+ .110
+ .006
— .068
+ .110
-.064
fa— 4
+ .018
+ .082
+ .065
+ .063
+ .021
+ .051
—
.064
+ .082
Oecond i‘actor loadine,3 and comaiunalities
A—1
A—2
A -3
A—4
B—1
15— 2
B— 3
B-4
Loading
.125
•268
.1 3 7
.401
.224
.182
.127
.191
Co.j.iunality
.0156
.0829
.0188
.1608
.0502
.0331
.0161
.0365
B-2
—
B-3
♦
B—4
—
Becond l Factor Residu ials
^-1
+
A—2
—
A-3
+
A-4
+
A-l
+ .202
+ .023
+.053
+ .005
+ .034
+ .202
+ .044
+ .O06
u-2
+ .023
+ .O6 0
-.034
+. 0 6 0
-.016
+ .041
+ .013
+ .027
a -3
+ .053
-.034
+ .098
-.076
+ .025
+ .098
+ .017
-.039
rt—4
+ .005
+ .060
-.076
+ .105
+ •0 38
-.105
+ .059
+ .014
H
1
td
+ .034
-.016
+ .025
+. 0 3 8
+ .083
+ .083
+ .022
-.022
B-2
+ .202
+. 0 4 1
+ .096
-.105
+ .083
+ •202
+ .091
+ .016
B-3
+ .044
+. 0 1 3
+ .017
+.059
+ .022
+ .091
+ .091
+ .088
fa—4
+ .0O 6
+ .027
-.039
+ .014
-.022
+ .016
+ .088
+ .088
B— 1
—
31
Second factor loadings and coxnmunalities
A— 1
A— 2
A-3
Loading
.362
.111
.0905
Communality
.1310
.0123
.0062
A—4
B—1
B-2
B-3
B-4
.C637
.157
.4 0 0
.271
.113
.0041
.0246
.I 60O
.0734
.0128
Comparison — Guessed & Computed CoiiLminaJ.ities
a
—
1
A—2
A— 3
a
—
4
B—1
B— 2
B—3
B—4
Guessed
.690
.4yO
.4ou
.Sue
.460
.SuC
.450
.390
Computed
.697
.5 0 2
.223
.898
.432
.720
.366
.283
V .0 2 8
+.080
+.084
+.107
Discrepancies -.007
- .0 1 2
+.175
- .0 9 8
Computations after computed coimminalities were substituted in
original correlation matrix.
Correlation Llatrix (computed cornmunalities)
A— 1
A —2
.46
A—1
A-3
A—4
B—1
B-2
B-3
E-4
.40
.69
.45
.36
.33
.34
.21
.49
.43
.37
.36
.39
.36
.26
.25
.20
.15
.46
.80
.34
.35
.31
.32
.31
.45
.30
^ 2
.19
a -2
.46
A-3
.40
.21
.*22
A—h
.69
.49
.36
L —1
.45
.43
.26
.46
B-2
.36
.37
.25
. SO
.31
B-3
.33
•36
.20
.34
.32
.45
E-4
.34
.39
.15
.35
.31
.30
.19
.28
First factor loadings and commonalities — second approximation
A—1
A—2
A-3
A—4
B—1
B-2
E—3
B-4
Loading
.749
.645
.414
.882
.597
.715
.514
•464
C o m aunal ity
.5610
.4160
.1714
.7779
.3564
.5112
.2642
.2153
First Factor Residuals — second approximation
A—1
—
A—2
+
A-3
—
A-4
—
B—1
+
B-2
—
B-3
+
B-4
+
A—1
+ .176
+ .023
+ .090
+ .029
-.003
-.176
+ .055
+.008
A—2
+ .023
+ .091
+ .057
+ .079
+ .045
+ .091
+ .028
+ .091
A-3
+ .090
+ .057
+ .090
-.005
-.013
-.046
+ .013
+ .042
A-A
+ .029
+ .079
-.005
+ .169
+ .067
+.169
+ .113
+ .059
B—1
-.003
+ .045
-.013
+ .067
+ .117
+ .117
+ .013
+ .033
B-2
-.176
+ .091
-.046
+ .169
+ .117
+.176
-.082
+ .032
B-3
+ .055
+ .028
+ .013
+ .113
+ .013
-.082
+ .11?
-.048
B-4
+ .008
+ .091
+ .042
+ .059
+ .033
+.032
-.048
+.091
Second factor loadings and comiaunali t ies — second approximation.
A—1
A—2
A-3
A-4
B—1
B-2
B— 3
B—4
Loading
.121
.303
.137
.407
.225
.168
.123
.185
Communality
.0146
0916
.0188
.1656
.0506
.0282
.0151
.0342
Second Factor Residuals — second approximation
A—1
+
A—2
A-3
+
A-4
B—1
B-2
+
B-3
—
—
—
—
B—4
+
A—1
+ .196
-.014
+.073
+ .020
+ .030
+ .196
-.040
-.014
A—2
-.014
+.044
+.016
+ .044
+ .023
-.040
+ .009
+ .035
A-3
+ .073
+ .016
-r°72
+ .061
+ .044
+.069
+ .004
+.017
A-4
+ .020
+ .044
+.061
+.101
-.025
+.101
+ .063
+.016
B—1
+ .030
+ .023
+ .044
-.025
+.°7?
+ .079
-.015
+.009
B-2
+ .196
-.040
+ .069
+.101
+.079
+ .196
-.103
-.001
B-3
-.040
+ .009
+ .004
+.063
-.015
-.103
+ .071
+ .071
B-4
-.014
+ .035
+ .017
+ .016
+.009
-.001
+ .071
The average
r
of the 28 sub— test intercorrelations is 0.369.
The standard error of this average
r
is 0.060.
From the table of
second factor residuals it may b e seen that these residuals range from
+O.I 96 to -0.103*
Pearson's test of goodness of fit was applied to
a distribution of these residuals.
It yielded a chi square of 5*725•
There were nine cells.
From Fisher's Tables Pearson's P— coefficient
was found to be 0.572.
That is, there are 572 chances in IOOO that
as poor a fit, or a poorer one, might be obtained in a chance distri­
bution.
It was decided that the second factor residuals, both with
respect to distribution and size, could have resulted by chance, and
that the extraction of a third factor would yield only spurious
results•
34
Comparison —
guessed and computed commonalities —
A—1
Discrepancies*
A-3
A-4
B-4
.900
.430
.720
.370
.280
-.576
-.508
-.190
-.943
-.407
-.539
-.279
-.242
+.023
+.181
+.091
+.031
.124
-.008
of
+.040
-.043
to the
the geometrical relationship of
factors.
Axis
Kj_ a n d K g r e p r e s e n t
theaxes before rotation.The broken lines,
labeled
i
and Kg represent the
this r o t a t i o n
the nature of
of this
position of the axes after
it appears
the
tyre which
to be a defensible
sub— t e s t s .
Vifhen t h e
The
did not ha v e
mechanical e x p e r i e n c e as w e l l
insight seems
elements
as
s u b j e c t A,
sigma
elements
factor
loadings
derived
these cases
procedure because
the
of
sub— test
solution calling
calling
from the
it w a s
logically evaluated in
Subject
engaged in the
zero or near zero
for
for m e c h a n i c a l
quite remote.
for m e c h a n i c a l
mechanical
in
experience and a
B,
insurance
experience and —0.7
score
For
obtained a score
of 1. 8 9
sigma
example,
o f 0.69
for m e c h a n i ­
in commerce and finance
obtained a
ior m e c h a n i c a l
a subjective evaluation of
the abilities
certain individuals.
a college graduate
business,
sub—test relationship
found that
a p r o f e s s o r o f a p p l i e d meclianics,
insight.
While
likelihood of constructing a
after r o t ation of axes w e r e used,
suggested were
rotation.
does not y i e l d a large p r o p o r t i o n o f
factor loadings,
cal
B-3
.230
the sub— testwith respect
1
B-2
.500
The following diagram represents
the position
B— 1
.700
Guessed
Computed
A—2
second approximation.
score of —1.40
insight.
the amount
for
In b o t h of
of the abilities
♦ <*-/
/T / & / O
rt" St
/v' 9 . < f /
/4 > re ^
W x e j
r o te * T / o r i
arff-cr- *3-&° r~o te r f t o n
6 ' e O / r 7 € ’/ ' / " / c a /
O f
^iorc?~o*~
r e j e /■«
Z o < 7 c ///7 < ^ r.
J'/or?
possessed b y the subjects, tallied closely with the above findings.
The following general equations were set u p to compute a 2—score for
any testee for the- two abilities identified b y factor one a n d factor
two:
0. 6 Beta — 0.25 Alpha - Phi
0.5 Alpha — 0*35 Beta - Theta
where,
Alpha * s u m of Z— scores on sub— tests
A— 1, A-3, A-4, B-3.
Beta =
sum of Z— scores o n sub— tests
B— 1, B-2, B-4, A— 2.
Phi *
the ability identified b y factor
one, called mechanical insight.
Theta = the ability identified b y factor
two, called mechanical experience.
APPEND IX H
Sub-test forking Drawings
The following drawings are detail working drawings fro m which
the parts of the sub— tests can be made.
These, together with the
assembly drawings of the sub—tests which are presented in the bod y of
the thesis, will b e suitable for use in the m anufacture of additional
sets of the Purdue Ldechanlcal Assembly Test.
BWMPg
— 1
■A-’ B earing *or r ^ a f i n q
M i m e r a u s j t rotation
horizon *a!. Dot h& qb* te e
a ssem b lies
B " B e a rin g *\.r cam a hn/r
fo r height :c e as»em clfes.
C ‘A lte r n a te b ea rin g for
i'm . h r h eig h t off assem h l/es.
D 'A m er can I ‘a rd o r a Oral
H ead W nt d act c /v *4• 41
\
Drillt;
E -F iat h'coo { op 5vf w
H -for $yb-tcstA-c, A-4,8 "3,H
is B f ;*or all otner^H is £ t
X "i/a r ia d c , t e e assem blies.
Y "Yar/'ot H, s■
_e a : *>err, bh'ee
!
i
(
fil
2.
t"
r~“i
a,_i~
---I
S)L ~ j Z p
-4«~ /
>
.
i
. /
**-, ,*’
A
ff *
s
t
'W#
.
‘. 't 1' Vtypical Bex Constlcficn.
* *v<.\ ?|:K; ’ I
S id es
B a t elite
t m v w m
RM.AI
SJB'TEGt
[
% COLLt-PM X.
nr
\ .... ■
.V ■
.
>■‘A
,t
t vklLL
n ’": i
if.
CnA»
Y-> Afi'
■■Vv"*
■ w.
I
C
!
PPL5S,
WT*?*
c
* ■v
'
- j / \
v_,y
-) 34*
.
I
OUT UNGTh !jr
*
I 1
T "; ■■«
»■ i
,
r^Ti
't-T-'r-*.,
"s' -v
•<
Vf
V
:::i- U ' l T
■-J
| ""^rrr ~(uc~3
; A^Jr/.VtO
4-4Y
,!.-.. /.n'iQO
r ~M - bb
•'. • - U;
:&i/I# /*&'■
,f?r
■.
i
i '
.**s ,
i■ i
•n
w p r t i m m
num. i m a m
1€
'V
f
,
M<
mm
I M
­
*h
RAL
p m v
PRESS* FIT
Afe. 14 GAU6C D m
PI
.3 .
...
Sr-
f -
'
u—
3j
ro#tnj
32F
-pm
■: u m
jlj
1
m b
/Vo, 4
STUL BAR STOCK
h
VS
'■ 'i l l
7
~v
m \S i
\
J
NO)
.
%j.'JKJ
'■^'*3
8g
.r«
fg
^
•v
.it
4«-;
v-a|
#»■>1
*(«
;;3
« *2
ii
^4
-<(■
.
^
S
i
'
'■A*
RAD/UC
RWKl"'
w v w m
4 tfMffSS
W.a-.'*j
no
*£•'«*
?t *
*?*
Bdsm m
Ms
pm nr
k s-tone-sn
\7 *■
■ 1
j
9
|~T.
nrfo .;~J4-. . ...—
—:jK
'A, I I
7V 1
y| V
..j^
„„ i
.:i r 7 .
i
rr^rrr
N a ltsmU:tS%u
m i M
i
u'X:
mss'w. '*■"•
-
m
4 /r
m
m
a m
100
$ '
/vi.
'
I:
i
) ssj V*
1
A ’A
...
«
m
stlll
COLLARS
$ a :5
ihi; u
m .
!‘:Ll4h
6PXX
\ jL w t
MAC
/
SRfc
44l'/j
fl'X
•• v
07
,
v,
A Vi
J/1
!
' " ~ I
1
S -'
T: I-'I'A!
■J.'
ki
t;5..I
"v '
jr „ /‘
■mi
jr :
f.T
*t
7
/
ir. 1 d
'//;M 7.v7
■
iVfj',.)'
■ ...A*A „■'
\.n\Hr ak<
xiOo.
1*01
pummnmr
-S .
~->j
%
i
j
p * « ‘
^
?:Wm> r'\t;\
,v y
4
'
,*•5?.'4l,;«
»v
t
P.:erfttSS
...»*
,*.,£l
.l
^yt.I,*-.*.
:
>i»Ji
l■
»-V*+»v
■ V
*/ " f
„& 1
,fiU->**..*«i
-i*? v’
s
< *
I
x%
• \ r - i
y A”
.V
■
■'
;t j
A,
A
1 ..r., ..
>*«*•*
M » » s -m
l$ ; * ‘
.
.
V
v #
»**>%
■ . UH if'l- .
r?4
I •i m u M l m
>
*«?
m
;•) *
;;
s m
y 4 4 .'•
<* #4«**#
*
iy ]'!
,y
,,\
a
l
V'4|;Aa ^
f.'rf' v. *,
%\ Yaffil;
7 ' }■/%'•
»t
’ *
W,
^
/
^04 '
;
£vh
'
re
0
& •
4
*
&
b ~ ~
A
J*i SA'i
<4
a®, m&im?-
'~i
i
p
a
K ,»H
LM3
a .z$s
t
•{ m o srcsi m - s t o u , 0 o
:!m M i
,
.*,
. '■■■■’ V
‘IS
>''** ■*
*,»‘ ' ’ v'VrTt
a,
'
V
r>
$fc£ i>/$£j
r•
Wi'4^1♦*
l'*‘ -A
,'
Vi
mift:mttmt&mtXi-P4^ .
Y f1
v?
Vt
rr
hr--- ^
L.U
i|
jhK J .>2
i A wnu o'feti dor ohcJ t'fcqjf.
■-
A A^ O n!:
.,
V:I
i t- ^
-0
.
J
' ■■
I
JTTCii.
•A
D "C
J ''.Cl
~'A '
.“ V
-
C
V
-
ill,
| ***
1*--■-/;?'-*“4 ■
...
D/d:Com-dr,
.
*,
hr.i-Dittt
/"" Hnyrl
A.
Do.5'
/
i"C 'i l
1
vo
j ,s «wrr*|}
eg . v
I,
..
Thumb
-6r:: 2 Reaa.
w « r u m m r
wiuumitmum
Ar
v■/
iPMAJ
.
mum. ".r -r iiiiiu
Mt
M
CM
M M MM
——. BSSUSB I
Xr
MLQ
Sftfi 8 A R j \ {
'
V
{,*
<
#
K
\t'(jkoo
/REQ.
*
4
fy.MS 6CFJ1
4
•<. >
1r
-
4
7! ►n
£*<,.
mm
Bn
H
/
.if'J
r j ^ t f
' p r e
s c je e
e
5 f ee i p i n
6$
fit ':
7/ //
a/- ■;r./.5'r‘
-
r/v. s.
/£K
•>
»
^
P t
Hearn
:- < \ h y b
C A P
« //w f M /ta w r
mifitmmimwm
/?yiMr
J?
id itot
" B E
■ M l
t
nimttu
mifti
T R U , I M
m
*
r' *
Wmfm
stseipin*
D rP s S f i t
i - f r x k ' B c i h ,
L C Q t
\i ;
y;
L
<?£***/
'S
B O S T O N
H 3 2 i i
IRi'Q.
B O S T O N
> R E Q
A
WJ
f a
hi, M V Z - 1 \/t Bet
fBoch 5c'Z fag.
J
5tee/
p in -p r e s s f i t
uQ.
SEA R*3
BOSTON M iTER C4C3 /££<?.
:uc.
h:
Sm
0 - - ^ -
Ml
>-
>» «
■<
4
/l
4
d-iS:
f»-*
C.
p S/a -x
ii Drill
//6
/£ y
t^T
'
"Tffl
?
‘
&
i
Drill
w s4"
COLLAR
8RAS5
IREq
r 0/
/6 Jl
-v»
■'55i
I ft,
z drj. 8t-
-C d h r
n
M
SEAR H
805TCN MiTER M
Eta.
T.:
f t
4*
>{
j
Tr
•f
»f
'■ •f
1y
.
'j
ft >,
1
'
OTTT ■
■
i+LlI
~;6DrHI
SEAR'S
8 0S T V iA O R M $ m I f i f y \
//err
m U E M K 1 0
m o F u t m t m a m
PMA.T
,
6MF£?
..
%B-TEITHDE1AM
M M
'\
‘.\S
.‘(S*!?W’i', V - V '
'2-.■/»?»
.*
\.S ' -- •
O' * '.i ' LX ■'
"
I “■ ■
•v
,f> '
• ' ”•..r- :
IS®
&G U CQUM-i
r iR ■
\$R
\
$ D R ILL
fi~
-
\,
,'
V!
•\viu
'//i" |
1■■AX
/
x
.
r
/#
A
** # *
H
(ft '
Q'D
TEEL
■0mm
r/;
d t
A
[J\K N /
J >' $h 1lLi )$r BAR S T j - j i
iXU ,■
iRFQ'D
;/
*
■■
I
i m t -JIMS
DISK C m - B R .
X
'
iREQ'EN
PINS - S T E E L
,
1
("i
V ~
Pj•
■~Cl/ •
/*
- ■ ii
►
m-ZREtD
.
m i
n
I
<3,MILD $7
'iV <
UA R 677,’
CA
//?£or
C A l AR-
ito; 'U
nmi
"Vi
Vi" &
jp'X-NiLU.
: A
rt t
3i: 1 I
1Li
I *■
, ■/
; V‘ • '
A S H
C AM-BR'
' R E Q'D
PAS-NEEL
it, T
: j.. ^
-i
J
Q<$k (ft!-'&-■
PM'^LL
nm
mi
m m
urn*wm
1RF.QV
■:
l'fIST'A-3DEI
«•
&
u.
jSj TJ]
;•«•*£V*V .*A
■m u £
—
-
1
A J-
%
V-
I
I' ■
-
£
L".„;.*
■'
fours
i
:'r;rs :
'(■
iJ'.;:
mm
im
*ir
W O
!\i
/VC3
A / 677: I CrciflK ttqndk f m
Set B$-ebttf£
0 ? Q ' 'A
nmv
1. A .
M U S T M DETAILS <
A
.
rmuu
MO
- XV. ^
^ffc [ < '1,* ’ '
—
1—
^
w n i N ^ i j lK j II. I ?
■»
i i n
* - x ; ^ > / “v" * V
1i i i i ii i M » i iii M i i i i M i i i * » ii <i ^
/
" • ' ' -“
\
ii| > | i i .[n i'i i 11 I I I
,
I W
.■*• ’X*lV ' ,
> 1. 1 . 1 . I ,. —
ajmEja*
■ ■ ■ !—
^'•g,A'
-AWWiSS'
* 1 ,*
%, f M S
.*.-&.>•£T *> A. " IJ1
.
\
*/* r ■ *
* \
■Mil'lliMiiimm
s
i*-
■/•
■rF'j'hJ;
/ifI
sd
1
'gRcidm
'V
. r
H
....
~
oH S - J^f —
M
1______ 4>T
U-./f -Z
2
i
75 „r
*
-•'/* ‘
hi
*8
**
nr~r
IiI
Z/io^ ih 4
I
*i m
^
so, jtod - t f y
KtJr!
}? * * f
T i ?
,, No5-40HC2
rs
x;
•v
I
7. £
Thum b
/A A,
Itv
.
'
X
n ' -■■V
- /
/f ' ■■
1 'f
..
Screw
' 1
I-..
.
A
mut m m
r.
mMtawLtt
Lr-*
J /f
P .M .A J .
/V ; 0
:r^/
/fa-, -'J'/r.-.^
M - T EST
m.
&
■
rmnu
si
A'i
%
Документ
Категория
Без категории
Просмотров
0
Размер файла
9 577 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа